v<v 
 
 <ks* 
 
 TVT* .A 
 
 
 ■ay ^ 
 
 * -—.% Pilfer \ / ^-^ 
 
 
 
 
 ^ 
 
 %\ .Astfi*^ «y^i:^ ^.ota*/*^ 
 
 
 
 
 
 <* 
 
 *°* 
 
 V "^ ^° ..., V"° 
 
 
 <^>**v 
 
 
 4 .«, 
 
 5? ^ 
 
 «*» *r 
 
 j>*^ 
 ^ ^ 
 
Digitized by the Internet Archive 
 in 2011 with funding from 
 The Library of Congress 
 
 http://www.archive.org/details/modernmethodsofwOOdavi 
 
MODERN METHODS OF WELDING 
 
MODERN 
 METHODS OF WELDING 
 
 AS APPLIED TO 
 
 WORKSHOP PRACTICE 
 
 DESCRIBING VARIOUS METHODS 
 
 OXY-ACETYLENE WELDING 
 OXY-HYDROGEN WELDING 
 LEAD BURNING 
 THERMIT WELDING 
 ELECTRIC ARC WELDING 
 ELECTRIC BUTT WELDING 
 
 ELECTRIC SEAM WELDING 
 ELECTRIC SPOT WELDING 
 MIRROR WELDING 
 CUTTING IRON AND STEEL 
 EYE-PROTECTION IN WELD- 
 ING OPERATIONS 
 
 AMERICAN METHODS 
 
 j/h. 
 
 BV 
 
 DA VIES 
 
 LEEDS TECHNICAL SCHOOL AND CONSULTING ENGINEER 
 
 NEW YORK 
 D. VAN NOSTRAND COMPANY 
 
 Eight Warren Street 
 1922 
 
"is 2,2,7 
 
 PRINTED IN GREXT BRITAIN BY BILLING AND S'dNS, LTD. 
 GUILDFORD AND ESHER 
 
 2. "3 
 
 ^ 
 
 Or 
 
 ^ 
 
01 
 
 CO 
 
 PREFACE 
 
 We live in a world of wonders. The life of each one of us is, of 
 necessity, so hemmed in by circumstances that none can see much 
 beyond the bounds of his own habitat. We recognise the progress 
 of an industry which comes within our own experience, but we know 
 little of those with which we are not in personal contact. 
 
 The pressman who clamps the plates on a modern lightning news- 
 paper press does not see anything very startling or interesting in the 
 work. The man who pulls levers in the pulpit of a great steelworks 
 is not apt to realise that there has been a marvellous advance in the 
 realm of manufacture. The attendant on a bottle-blowing machine 
 has learned to take his work as a matter of course. It is found 
 through the entire list of trades and occupations. 
 
 Yet each of these men is at times impressed by the remarkable 
 advances made in some industry other than his own, because such 
 knowledge comes to him, as it were, suddenly, not by the almost 
 imperceptible movement which marks progress in work that is 
 familiar. 
 
 The means which have brought about industrial development 
 are worth studying. No armed warrior ever sprang full-grown 
 from his cradle; no giant industry has ever come into being in a 
 year or decade. It takes time to develop the machinery and acquaint 
 the world with the advantages of a new aid to manufacture, to 
 commerce, to civilisation, or to human comfort. He who reads this 
 book can hardly fail to be impressed with the idea that no man 
 can measure the possibilities of industrial growth. Who can say 
 that at the close of the twentieth century " Darkest Africa " may 
 not be under-selling us in our home markets? Who can be sure that 
 with the development of China and the East, there may not come 
 supremacy in industry before which our light shall pale ? To-day 
 the industry is exceptional in which there has not been an entire 
 alteration and renewal in the machinery within the last few years. 
 
 The tale of manufacturing progress is one not half told, one which 
 never can be told in full, because it grows faster than the ability 
 
vi PREFACE 
 
 to record its development. In every vocation, in every city of the 
 globe, are geniuses studying how to advance the lines of work in which 
 they are engaged. Every year the standards that win success are 
 set higher, yet every year witnesses increasing gains and greater 
 triumphs. 
 
 Electricity is believed to pervade the universe. Astronomers 
 see evidence of its action in the sun, in the stars, in the comets. 
 Its properties are so varied, it affects substances so differently, 
 that it is safe to say that we have as yet learned but a fraction of 
 what mankind is destined to know about this wonderful thing. 
 Because it so readily lends itself to the transmission of energy, we 
 think of it as a source of power, whereas really it is but a means of 
 transmitting power, like the endless leather belts commonly used 
 for driving machinery. 
 
 The man who thinks he will read up a little on electricity is some- 
 times very much disappointed because he cannot learn at the out- 
 set, in a little primer, just what electricity is, and so advance step 
 by step to a full knowledge of the subject. But there is no help for 
 it. The operator of electricity to-day must begin, as did those who 
 came before him, at the other end of the problem, and learn how 
 electricity acts and what it does. After a time he will acquire a 
 notion of things which will satisfy his craving for knowledge, and 
 will cease to bother much about the theory. 
 
 Operators and others who follow the instructions in this book will 
 soon be convinced of the great importance of welding processes to 
 the future manufacturing and industrial world. It is the simplest 
 possible axiom, when we stop to think (though few people ever do 
 stop to think), that the only way in the long run for labour as a 
 whole to get more wealth is for it to create more wealth ; the only 
 way to create more wealth is to increase productivity of labour. 
 
 The field for the further application of welding is enormous; 
 but this further application is being delayed by lack of com- 
 plete knowledge of the art, the utterly confusing, and, in many 
 cases, diametrically opposed claims of competing interests. There 
 is needed a cultivation of the co-operation spirit which will permit 
 a frank, open discussion of the merits of the different processes, 
 so that a reasonable agreement as to those merits may be reached. 
 If there are prospective users of welding who are in doubt as to 
 whether they should use gas or electric welding, or neither, can it be 
 supposed that their confidence in any process will be enhanced by 
 hearing its advocates claim that it is the only safe and economical 
 one ? I am not setting forth impracticable ideals, but rather 
 
PREFACE vii 
 
 common-sense principles, already found successful in many business 
 fields, the application of which is bound to yield the best results 
 for all concerned. 
 
 It would be difficult to suggest a branch of the applied arts which 
 has advanced more rapidly in recent years than that of electric and 
 oxy-acetylene welding. Both processes gained status in the war; 
 and, although some of the more extreme manifestations of their 
 possibilities which the war encouraged are little likely to be paralleled 
 in the early days of peace, the methods have now won for themselves 
 a definite place in the shop routine. They have established their 
 ability to tackle certain classes of work in an economical and satis- 
 factory way. That both methods are destined to advance in useful- 
 ness, alike in extension and intension, there is no question; and, 
 although they are in the hands of specialist firms, the general shop 
 manager of the not far distant future is likely to find that he is 
 expected to be able to apply them. Moreover, not only the manager, 
 but others, both the engineers and the " semi-skilled," are likely to 
 find interest and profit in these processes. There is probably no 
 other recent improvement in applied mechanics which has received 
 the scientific study devoted to autogenous welding. The result is 
 that its practice has become an art to which one must give intelligent 
 and well-directed study if one would avail oneself of its uses. 
 
 The universal application of this process, in every branch of 
 industry where metals are employed, makes inevitable a great de- 
 mand for capable operators. It is to the advantage of all concerned 
 that operators should understand their work and their tools, that 
 they should be able to apply intelligently the principles involved. 
 
INTRODUCTION 
 
 In presenting this book to the welding industry, I may say that I 
 have devoted my whole time to the welding processes and the 
 materials used, and they are all concisely described in their various 
 chapters. If the readers will give attention to the following pages, 
 they will find many points that will help them in their studies and 
 guide them to a knowledge of the processes. In the course of 
 eighteen years' experience of the welding industry I have found that 
 such a book as the present is badly needed. I now make an en- 
 deavour to supply the deficiency. It is my desire to do all in my 
 power to raise the status of welders in this country; and this will 
 be best achieved if they can be induced to pursue a course which 
 will make them proficient. 
 
 Some time ago an effort was made to establish a system of certifi- 
 cates of proficiency for operators, but unity between the associa- 
 tions interested is not yet sufficient to allow this. 
 
 In the comparatively limited space available for my purpose, 
 I have attempted to give a clear and consecutive description of the 
 principles upon which an industry of unsurpassed importance is 
 based. With the object of accomplishing my task, however, in a 
 manner at once agreeable and instructive, I have now and again 
 departed from the general plan to dwell on some particular point. 
 
 While sensible of the defects in my book, I venture to hope that 
 to those practically engaged or interested in the conduct of numerous 
 processes covered by its title it may prove to be of service. 
 
 I take this opportunity of acknowledging my gratitude for 
 assistance from Messrs. British Oxygen Company, Ltd., Messrs. 
 Charles Bingham and Company, Ltd., Messrs. Leeds and Butter- 
 field, Messrs. British Insulated and Helsby Cables, Ltd., and H. M. 
 Hobart, President of the American Welding Society. 
 
CONTENTS 
 
 CHAPTER PAGES 
 
 PREFACE ------- V 
 
 INTRODUCTION ------ ix 
 
 I. WELDING IN GENERAL ----- 1-5 
 
 Concise description. Articles welded. 
 
 II. LEAD BURNING - - - - - - 6-12 
 
 Description of the gases. Samples of welds. 
 
 III. MANUFACTURE OF OXYGEN - - - - 13-16 
 
 Describing Brin process. Linde liquid air. Process described. 
 
 IV. MANUFACTURE OF HYDROGEN - - - - 17-21 
 
 Description of process. General details. 
 
 V. MANUFACTURE OF CARBIDE ... - 22-26 
 
 History. Starting-point of the acetylene industry. Principal 
 materials for carbide. 
 
 VI. ACETYLENE ...... 27-34 
 
 Physical properties. Weight of water to weight of carbide. 
 Impurities. Purifying materials. 
 
 VII. OXYGEN CYLINDERS ----- 35-39 
 
 Manufacture of cylinders. Regulations. 
 
 VIII. ACETYLENE GENERATORS - - - - - 40-45 
 
 General construction. Polymerisation. Working of generator 
 described. General rules. 
 
 IX. OXYGEN REGULATORS ----- 46-48 
 
 Description of regulator. Special coupling. 
 
 X. REGULATIONS ------ 49-51 
 
 Regulations for generators. 
 
 XI. BLOWPIPES ...... 52-64 
 
 Pattern and consumption. Series of blowpipes. Testing the 
 gases. General rules. 
 
 XII. FLEXIBLE TUBING ..... 65-66 
 
 Quality of rubber tubing. 
 
xii CONTENTS 
 
 CHAPTER PAGES 
 
 XIII. SAFETY VALVES ..... 67-71 
 
 Descriptive notes. Choice of safety valve. 
 
 XIV. PURIFIERS ...... 72-77 
 
 Remarks on design. Testing the purifier. The Keppler test. 
 
 XV. SELECTION AND INSTALLATION - - - 78-82 
 
 Generators described. Conditions a generator should fulfil. 
 Order of installation. 
 
 XVI. METHODS OF WELDING .... 83-92 
 
 Preparations prior to welding. What causes adhesion. 
 Test-pieces for operators. 
 
 XVII. PREPARATION OF WELDS .... 93-99 
 
 Bevelling. Defective welding illustrated. Phenomenon of 
 expansion. 
 
 XVIII. WELDING TABLES ..... 100-103 
 
 Construction. Illustration of canting tables. 
 
 XIX. FURNACES FOR HEATING ... - 104-108 
 
 Description of furnace. Best method of preheating. 
 Expansion and contraction. 
 
 XX. IRON AND STEEL ----- 109-119 
 
 Difference between iron and steel. Physical properties. 
 Pure iron. Interposition of oxide. 
 
 XXI. CAST IRON ------ 120-L31 
 
 The strength and solidity. Action on cooling. White and 
 grey iron. Difficulties of expansion. 
 
 XXII. DISSOLVED ACETYLENE ... - 132-134 
 
 Brief description. Regulations. Correct tip of flame. 
 
 XXIII. CUTTING IRON AND STEEL ... - 135-153 
 
 Illustration of cutter. Oxidation of iron and steel under 
 oxygen. Heat value of carbo -hydrogen. 
 
 XXIV. THERMIT WELDING ----- 154-159 
 
 Preparation. Thermit required for welding. Composition 
 of thermit. 
 
 XXV. PROPERTIES OF PRINCIPAL NON-FERROUS METALS - 160-163 
 Important properties. Table of melting-points. 
 
 XXVI. DELTA METALS ..... 164-167 
 
 Table of metals. Important to operators. 
 
 XXVII. ALUMINIUM - - - - - -168-176 
 
 Comparative cost. Melting-points of the metal and its 
 oxide. Alloy mixture. Points on welding. 
 
 xxviii. copper ...... 177-183 
 
 Principal varieties. Special welding-rods. Failures in 
 welding. 
 
CONTENTS xiii 
 
 CHAPTER PAGES 
 
 XXIX. BRONZE ------ 184-186 
 
 Bronze welding. Power of the blowpipe. 
 
 XXX. BRASS .--... 187-190 
 
 Physical properties. Melting-points. 
 
 XXXI. AMERICAN METHODS - 191-199 
 
 Description of what is being done. Welding machines. 
 Medium-pressure generator. Cutting machines. Non- 
 flash blowpipe. 
 
 XXXII. THE METALLURGY OF ARC WELDING - - 200-205 
 
 Metallurgy of steel. Gas-holes in electric welds. Im- 
 purities. 
 
 XXXIII. BRIEF DESCRIPTION OF ELECTRIC WELDING - 206-210 
 
 Spot, butt, seam arc detailed. Relative cost. Rivets 
 versus welds. 
 
 XXXIV. ELECTRIC ARC WELDING ... - 211-226 
 
 General processes. Quasi-arc process. Welding with 
 metallic electrodes. Practical notes. 
 
 XXXV. SPOT WELDING ----- 227-235 
 
 Sheets to be welded. Articles that can be spot welded. 
 
 XXXVI. ELECTRIC BUTT WELDING - - - - 236-243 
 
 Butt- welding processes. Electric current no effect on 
 weld. Chain welding. 
 
 XXXVII. ELECTRIC SEAM WELDING - - - - 244-246 
 
 Welding sheet brass. Working operation. 
 
 XXXVIII. EYE-PROTECTION IN IRON WELDING OPERATIONS - 247-257 
 Remarks on types of glass. Proper selection of colour. 
 Summarising. 
 
 XXXIX. MIRROR WELDING ----- 258-263 
 Describing the process. Showing how mirror welding can 
 be accomplished. Welding a boiler. 
 
LIST OF ILLUSTRATIONS 
 
 FIG. PAGE 
 
 1. CUTTING BOILER FLANGES AUTOMATICALLY WITH A " RADIOGRAPH " - 5 
 
 2. INJECTOR BLOWPIPE ....... 7 
 
 3. SHOULDER TAPS FOR USE WITH LEAD-BURNING BLOWPIPE - - 7 
 
 4. OXYGEN REGULATOR ....... 8 
 
 5. LEAD-BURNED JOINT ....... 9 
 
 6. LEAD- BURNED JOINT ....... 9 
 
 7. MOULDING TOOL FOR HEAVY VERTICAL WELDING - * - - 10 
 
 8. 4,000-UNIT GENERATORS FOR PRODUCING OXYGEN AND HYDROGEN - 20 
 
 9. SECTION OF AN OXYGEN CYLINDER - - - - - 39 
 
 10. 10-CWT. CARBIDE-TO-WATER GENERATOR - - - - 42 
 
 11. DAVIS - BOURNONVILLE 50 POUNDS CAPACITY MEDIUM - PRESSURE 
 
 GENERATOR ........ 43 
 
 12. DAVIS - BOURNONVILLE 100 POUNDS CAPACITY MEDIUM-PRESSURE 
 
 GENERATOR ----.... 44 
 
 13. OXYGEN REGULATORS: 1 TWO GAUGES, 1 ONE GAUGE, 1 NO GAUGE - 46 
 
 14. DOUBLE CYLINDER CONNECTOR - - - - - - 47 
 
 15. AIRTIGHT CARBIDE CHAMBER, " ATOX " TYPE - - - - 50 
 
 16. FOUCHE BLOWPIPE, SECTION AND ELEVATION - - - 52 
 
 17. UNIVERSAL TYPE BLOWPIPE, HEAD AT SPECIAL ANGLE - - 53 
 
 18. UNIVERSAL SINGLE TYPE BLOWPIPE - - - - 53 
 
 19. UNIVERSAL MULTIPLE TYPE BLOWPIPE, SMALL SIZE - - 54 
 
 20. UNIVERSAL MULTIPLE TYPE BLOWPIPE, LARGE SIZE - - - 54 
 
 21. ENDAZZLE BLOWPIPE, SINGLE TIP PATTERN - - - - 55 
 
 22. ENDAZZLE BLOWPIPE, MULTIPLE TIP PATTERN - - - - 56 
 
 23. OSBORNE BLOWPIPES, FOUR DIFFERENT TYPES - - - - 57 
 
 24. DAVIS-BOURNONVILLE SMALL SET OF INTERCHANGEABLE TIPS- - 58 
 
 25. DAVIS-BOURNONVILLE LARGE SET OF INTERCHANGEABLE TIPS - - 59 
 
 26. SHOWING THE CORRECT NEUTRAL FLAME - - - - 63 
 
 27. SAFETY VALVE, ELEVATION AND SECTION - - - - 68 
 
 28. STANDARD SECTIONAL TYPE SAFETY VALVE - - - - 69 
 
 29. PURIFIER, SHOWING SECTION AND ELEVATION WITH PURIFYING MATERIAL 74 
 
 30. " ATOX " PURIFIER - - - - - - - 75 
 
 31. BUTT JOINT ....---- 85 
 
 32. INDENTS, NOT SUFFICIENT WELDING-ROD ON - - - - 85 
 
 XV 
 
XVI 
 
 LIST OF ILLUSTRATIONS 
 
 FIG. 
 
 33. ROUND BAB BEVEL, CHISEL POINTS 
 
 34. TUBE BEVEL ...... 
 
 35. FRACTURE AFTER BENDING .... 
 
 36. LACK OE PENETRATION ----- 
 
 37. SINGLE BEVELLED JOINT .... 
 
 38. ADHESION, BAD WELD ..... 
 
 39. NOT FULLY PENETRATED, FIRST STARTING OF FRACTURE 
 
 40. NOT PENETRATED THROUGH .... 
 
 41. BENDING A TEST- PIECE IN A VICE 
 
 42. FLAT BAR DOUBLE BEVEL .... 
 
 43. ANGLE IRON, BEVELLED ONE SIDE 
 
 44. NOT PENETRATED ..... 
 
 45. SPACE SHOWN UNWELDED .... 
 
 46. MACHINE OPERATOR WELDING SEAMS 116 INCHES LONG 
 
 47. CASTING BROKEN, SET AND CRAMPED, AND WELDED - 
 
 48. PROPER SET FOR WELDING PIPES ... 
 
 49. WELDED CROSS IN PIPE .... 
 
 50. SHOWING CONSTRUCTION OF PIPE 
 
 51. SHOWING PIPE AT 45 DEGREES, WELD AND BEVEL 
 
 52. MILD STEEL OPERATOR'S TABLE - - - 
 
 53. ADJUSTABLE WELDING TABLE WITH VICE ATTACHMENT 
 
 54. OPERATING TABLE WITH FIRE-BRICKS 
 
 55. TABLE FOR A NUMBER OF OPERATORS - 
 
 56. TILTING TABLE ...... 
 
 57- A KEROSENE PREHEATING TORCH 
 
 58. GAS-HEATED PREHEATING OVEN - - - - 
 
 59. ASBESTOS SCREEN FOR COVERING CASTINGS 
 
 60. LIGHT LIFTING CRANE - - , . 
 
 61. TENSIONAL TESTS - - - - 
 
 62. CAST-STEEL TRAMWAY GEAR CASE WELDED 
 
 63. CAST-IRON PRESS SIDE BROKEN - - - - 
 
 64. CAST-IRON PRESS SIDE WELDED 
 
 65. TWO-CYLINDER MOTOR ENGINE BROKEN 
 
 66. TWO-CYLINDER MOTOR ENGINE WELDED 
 
 67. BROKEN WATER PUMP ..... 
 
 68. PERFECT NEUTRAL FLAME .... 
 
 69. HAND-CUTTING BLOWPIPE .... 
 
 70. THE " RADIOGRAPH " CUTTING STEEL PLATE 
 
 71. COUPLED CYLINDERS, FOR CONTINUOUS WORK - 
 
 72. CUTTING ROUND THE FLANGE OF THE END OF A BOILER 
 
 73. CIRCULAR CUTS IN STEEL PLATES 2£ INCHES THICK 
 
 74. CUTTING A CAST-STEEL TURBINE ROTOR 9 INCHES THICK 
 
 75. FELLING A STACK WITH AN OXYGEN CUTTER 
 
 87 
 
 87 
 88 
 88 
 89 
 89 
 90 
 90 
 91 
 94 
 94 
 95 
 95 
 96 
 97 
 98 
 98 
 98 
 98 
 100 
 101 
 102 
 102 
 103 
 104 
 105 
 106 
 107 
 117 
 121 
 122 
 122 
 129 
 130 
 131 
 133 
 136 
 139 
 140 
 141 
 142 
 144 
 146 
 
LIST OF ILLUSTRATIONS 
 
 xvn 
 
 FIG. 
 
 76. 
 
 77. 
 78. 
 79. 
 80. 
 81. 
 82. 
 83. 
 84. 
 83. 
 86. 
 87. 
 
 90. 
 
 91. 
 
 92. 
 
 93. 
 
 94. 
 
 95. 
 
 96. 
 
 97. 
 
 98. 
 
 99. 
 100. 
 101 
 
 02. 
 103. 
 104. 
 105. 
 106. 
 107. 
 108. 
 109. 
 110. 
 111. 
 112. 
 113. 
 114. 
 115. 
 116. 
 117. 
 118. 
 
 PAGE 
 
 THE lA OXYGRAPH, AUTOMATIC PANTOGRAPH CUTTING MACHINE - 147 
 
 OXYGRAPH TRACER WHEEL, SWIVEL STANDARD .... 148 
 
 OXYGRAPH MACHINE CUTTING TORCH WITH MOTOR CONTROL SWITCH - 149 
 
 SMALL SOLID END CONNECTING-ROD AND BILLET - - - 150 
 
 LEATHER- CUTTING PUNCH FOR SHOE MANUFACTURE - - - 150 
 
 WRENCH TRIMMING DIE ROUGHED OUT ON OXYGRAPH - - - 151 
 
 VIEWS OF THREE DIES CUT FROM 110-POINT CARBON STEEL - - 152 
 
 NEW TEETH PUT IN WHEEL BY " THERMIT " WELDING - - 154 
 
 THERMIT- WELDED LOCO FRAME - - - - - - 155 
 
 THERMIT-WELDED ROCK CRUSHER - - - - - 158 
 
 FRACTURED ALUMINIUM GEAR CASE - - - - - 171 
 
 FRACTURED ALUMINIUM- ALLOY GEAR CASE - - - - 175 
 
 ALUMINIUM- ALLOY GEAR CASE REPAIRED - - - - 176 
 
 SECTION OF A COPPER WELD - - - - - - 179 
 
 MICROPHOTOGRAPHS FROM THE REGION OF A WELD - - - 182 
 
 THE DUOGRAPH WELDING MACHINE ..... 192 
 
 SHOWING DIFFERENT OPERATIONS OF THE DUOGRAPH - - - 193 
 
 MEDIUM-PRESSURE GENERATOR, 200 POUNDS CAPACITY - - - 195 
 
 PART SECTION, NON-FLASH BLOWPIPE ..... 196 
 
 AIR-GAS PREHEATING TORCH - - • - - - 197 
 
 WELDING TORCH TIP DIRECTLY ON METAL .... 197 
 
 WELDING TORCH TIP DIRECTLY AGAINST BRICK - - - 198 
 
 DRILLING A HOLE THROUGH 5-INCH AXLE - - - - 199 
 
 SHOWING AREAS AT HIGH MAGNIFICATION - - - - 201 
 
 SHOWING AREAS AT HIGH MAGNIFICATION .... 201 
 
 ELECTROLYTIC IRON ....--- 202 
 
 ELECTROLYTIC IRON NITROGENISED BY ANNEALING - - - 202 
 
 EDGE OF WELD MADE WITH COVERED ELECTRODE - - - 203 
 
 UNANNEALED WELD SECTION ...... 203 
 
 ADHESION IN WELD ....... 204 
 
 SLAG ENCLOSED IN WELD ...... 204 
 
 USE OF METAL ELECTRODE IN WELDING STEEL BANDS - - 217 
 
 USE OF CARBON ELECTRODE WITH METAL FILLER - - - 218 
 
 SEAM PREPARED FOR HAND WELDING WITH CARBON ELECTRODE - 219 
 
 COMPLETE UNIT FOR ELECTRIC ARC WELDING - - - - 221 
 
 AUXILIARY PANEL ....... 222 
 
 PRESCOT SPOT WELDER ------- 230 
 
 SPOT-WELD FLAT BARS TO CORRUGATED SHEETS - - - 231 
 
 PIECE OF ANGLE IRON TO FLAT PLATE - - - - - 231 
 
 ROUND IRON TO ANGLE IRON ...... 232 
 
 BUTT-WELDING TOOL STEEL ...--- 239 
 
 FLASH AND UPSET WELD ...... 240 
 
 PRESCOT BUTT WELDER ------- 241 
 
xviii LIST OF ILLUSTRATIONS 
 
 FIG. PAGE 
 
 119. BUTT WELDER MAKING CHAINS - - 242 
 
 120. ELECTRIC SEAM-WELDING MACHINE ----- 245 
 
 121. SPECTRUM OF A MAZDA LAMP ---.._ 248 
 
 122. PFUND GOLD GLASS GOGGLES ----._ 248 
 
 123. POPULAR FORM OF HELMET WITH CIRCULAR WINDOW - - - 249 
 
 124. WELDER'S HAND SHIELD ...... 250 
 
 125. ALUMINIUM HELMET, FRONT AND BACK VIEWS - - - . 251 
 
 126. SUNDRY SPECTRA 6 ...... 252 
 
 127. SUNDRY SPECTRA 7 ...... 253 
 
 128. FIBRE GOGGLES -------. 254 
 
 129. TWO PAIRS OF GOGGLES FITTED WITH ESSENTIALITE AMBER LENSES - 257 
 
 130. PRINCIPLE OF MIRROR WELDING ..... 259 
 
 131. MIRROR WELDING AS APPLIED TO THE PIPES .... 260 
 
 132. INTERNAL WELDING OF A BOILER ..... 261 
 
MODERN 
 METHODS OF WELDING 
 
 CHAPTER I 
 WELDING IN GENERAL 
 
 It is well known to everyone who takes an interest in welding and 
 welding processes that the existing opinion as to the value of the 
 processes and the practical results obtained is in a state of uncer- 
 tainty. The chief ground for this uncertainty lies in the fact that 
 these new processes have only been introduced recently into indus- 
 trial practice, and rest entirely on an empirical basis. Although 
 oxy- acetylene welding is now extensively used, and is of great 
 theoretical and practical interest, it has never been made the object 
 of systematic and exhaustive research. 
 
 The author has had eighteen years' experience of welding, and has 
 made exhaustive tests and long studies, not only of what is being 
 done in this country, but also of the progress made in the United 
 States and Germany. The latter country is far more advanced 
 than Great Britain. The Germans have carried out systematic 
 and exhaustive researches. Their operators are scientifically 
 trained, are taught metallurgy and chemistry, including the chemi- 
 cal compositions and melting-points of all metals and oxides, make 
 test-pieces for experimenting with the chemical and mechanical 
 tests, and employ microscopic and macroscopic examinations, both 
 of the melted zone and the neighbouring parts. Their welding, as 
 a rule, is very neat, as they are not allowed to execute commercial 
 work until they have become fully proficient. 
 
 The introduction of oxy-acetylene welding has opened up an 
 enormous field, in which any metal can be dealt with, and such an 
 article as a cracked motor frame or cylinder can be rapidly welded. 
 In these directions there seems to be ample scope for the applica- 
 tion of engineering skill, and recent developments have shown that 
 it is difficult to put a limit to the purpose to which engineers may 
 yet apply the process. 
 
 To-day the business has grown beyond the limits of personal 
 supervision. The necessity for organised instruction of operators 
 is becoming more and more obvious in the interest of both work- 
 
 l 
 
2 MODERN METHODS OF WELDING 
 
 men and employer. Several welding schools have now been started 
 in various centres about the country, whence a stream of qualified 
 welders is already beginning to flow to the workshops, where most 
 of them are able to turn their training to practical use. 
 
 They teach the operator under practical conditions the right 
 flame for different work, the principles on which the blowpipe is 
 constructed, the way to handle it, and a variety of technical 
 and theoretical points, which are always useful to him in his sub- 
 sequent career. He is also taught thoroughly the construction, 
 working, and maintenance of the plant. 
 
 It is the operator of to-day, well instructed in the points, whom 
 we hope to find the professional welder of to-morrow. The time 
 is not far off when employers will refuse to engage a welder unless 
 he can produce the certificate of proficiency. This cannot be 
 obtained unless the operator possesses thorough knowledge and 
 practical experience of the process. 
 
 The author has undertaken many investigations in this process. 
 The oxy-acetylene method, when properly worked, possesses 
 numerous marked advantages. In the first place, the operating 
 flame can easily be controlled, and the temperature attained at 
 various zones can be readily regulated. Secondly, the work can 
 be easily accomplished, owing to the high temperatures reached 
 (3,600° C), and the appliances are convenient to handle and reliable 
 in operation. 
 
 The most important conditions for securing good results are — 
 
 1. The use of the purest acetylene possible. 
 
 2. The use of a blowpipe so designed as to ensure accurate 
 adjustment in the proportion of the mixed gases and to secure their 
 exit at a velocity capable of keeping the metal sufficiently fluid 
 without the melting flame being too rigid. 
 
 3. The use of an absolutely pure welding-rod. 
 
 4. The provision of an absolute neutral zone in the melting flame, 
 neither oxidising nor reducing. 
 
 5. The edges must be free from all impurities, and, if over 
 -£$ inch thick, must be bevelled. 
 
 6. The use of deoxidising agents eliminating the oxides, in view 
 of unavoidable oxidation of the metal subject to the melting process. 
 It is necessary to bear in mind the relation between the melting- 
 points of the oxides and of the metal itself, which is a most important 
 matter. 
 
 7. Rapidity in melting, in order to avoid excessive heating, 
 which not only alters and deteriorates the original structure of 
 
WELDING IN GENERAL , 3 
 
 the metal, but would even favour the occlusion of the gases (particu- 
 larly hydrogen) and so occasion the formation of blowholes in the 
 melted zone. 
 
 In addition to these considerations, care should be taken that 
 no sudden cooling occurs. The conditions may have to be modified 
 on account of the conductivity and special dilation of the material, 
 as well as in relation to the thickness, size, and shape of pieces 
 operated upon. 
 
 Eor those who are familiar with this process of welding and cut- 
 ting it is not difficult to appreciate its varied applications. The 
 ease and rapidity with which experienced welders can carry out 
 repairs in situ, and the portability of the plants, make the process 
 valuable, if not indispensable. The service rendered by it in many 
 workshops, where the welding of articles of all kinds is a daily 
 necessity, is calculable. 
 
 The oxy-acetylene process occupies a leading place in all aero- 
 plane and airship industries. It is used with advantage in welding 
 sheet steel stampings, cylinders, aluminium crank cases and 
 machinery parts, steel tubes, stamped steel water-jackets for cylin- 
 ders, broken cast iron. Moreover, for cutting iron and steel this 
 process has no rival whatever. It will cut wrought iron or steel 
 plate 20 inches thick. The flame has been applied to the case- 
 hardening of steel, and some firms are using this on a large scale. 
 It is well known that the flame containing an excess of acetylene 
 is a very energetic carboniser. 
 
 This process may be employed on any class of work. It will weld 
 30-gauge steel or lj-inch boiler plates, cut mild steel up to 20 inches 
 thick; weld any commercial metal, such as cast iron, aluminium, 
 copper, bronze, zinc, lead, delta metal. In the repair of broken 
 machinery and parts it is always above its rivals; repairs are exe- 
 cuted quickly and can be done without dismantling in many cases. 
 It can be used in the manufacture of safes and tanks, in the jointing 
 of pipes, steam superheaters, casks, artistic ironwork, in adding 
 metal to parts worn by friction, filling up holes or parts of new 
 structure cut away in error, welding of tool steel to wrought- 
 iron bars, and welding of copper or brass tubes. The flame can be 
 used for preheating and for hammering and annealing after welding, 
 thereby ensuring a soft metal, a method not practicable in the electric 
 process. 
 
 In the welding of light sheet steels with 24-gauge metal, 45 feet 
 per hour can be welded, with a consumption of only 4 feet of oxygen 
 and 3 feet of acetylene. On the other hand, when the metal reaches 
 
4 MODERN METHODS OP WELDING 
 
 | inch thick, electric welding has the advantage both in speed and 
 cost. 
 
 When the process of acetylene welding was first introduced, its 
 apparent simplicity led many engineers wrongly to assume that 
 welding appliances might be regarded as general workshop tools, 
 which any inexperienced but handy man could operate with success. 
 Consequently much work was condemned wholesale because of the 
 defects in the weld. The author would emphasise that this is not 
 the fault of the process, but of inefficient workmen. 
 
 It is estimated that there were, during the recent war, 33,000 
 employed in this country in welding processes, of whom 25,000 
 entered the field during the war. Of the total number, 90 per cent, 
 are not fully skilled — that is, they are incapable of executing satis- 
 factory welds on all metals, being mostly employed on sheet steel. 
 The impetus that has been given under war conditions should stimu- 
 late employers to investigate and exploit this revolutionary process, 
 the possibilities of which have no obvious limits. 
 
 In the shipbuilding trade this process can be utilised very much 
 more than at present — for instance, in making knee brackets, 
 stays, and frames. These can all be cut and welded by blowpipes, 
 with present costs reduced 50 per cent, and output increased. 
 A blowpipe only requires one man ; but an anglesmith, when welding 
 a knee bracket, requires two or three assistants. Most shipyards 
 have plants, but they are not utilised to advantage. 
 
 In all Avelding it is most important that the work should be ade- 
 quately prepared before commencing to weld, as all time spent in 
 this way is amply repaid afterwards in the easier execution, and 
 also by the homogeneous nature of the weld. It is, however, a 
 subject on which it is impossible to lay down any hard-and-fast 
 rules, the varying nature of the work accomplished making it im- 
 possible to do so. The general principles obtained in the best prac- 
 tice point out that the line of weld must be opened out — that is, 
 the two edges must be bevelled to an angle of 45 degrees, to make 
 certain that the weld is well penetrated, not merely sealed over, and 
 to strengthen the weld by increasing the surface of contact. 
 
 One of the most important things to do in the preparation is to 
 arrange the pieces to be welded in such a position that there shall 
 be no deformation, breaks, or cracks, or internal strains, and that 
 they may be linable at the conclusion of the operation. This is a 
 point in which the skill and experience of the operator are revealed, 
 as there are no rules to guide him, and upon any work but that of 
 the simplest character failure to grasp and apply the laws of expan- 
 
WELDING IN GENERAL 5 
 
 sion and contraction means the partial or total ruin of the work. 
 It is impossible to control expansion and contraction by physical 
 force, so the only way to prevent disastrous results is to foresee 
 the probable direction and extent of the phenomena, and nullify 
 the effects by preheating certain parts of the work, either by the 
 blowpipe or the welder's furnace. 
 
 It is good practice to raise the temperature to nearly red heat 
 in the furnace and weld, then allow to cool slowly and uniformly 
 
 Fig. 1. — The Radiograph, used for Circular Cutting, Automatically, as 
 shown, is Cutting Annular Rings 31 Inches Diameter by 6 Inches 
 Thick at a Speed of 7 Inches per Minute. 
 
 Note the clean cut and accuracy of the rings. 
 
 in sand or asbestos; but all cold currents of air must be avoided. 
 All castings should be preheated bodily. This is a great advantage, 
 for not only does it save gases, but it prevents any irregular expan- 
 sion, and hence no fractures. 
 
 The cutting of iron and steel by the oxy-acetylene flame is being 
 very extensively used. This involves the use of a blowpipe of a 
 different design, which provides for the oxy-acetylene flame and an 
 auxiliary jet of pure oxygen to be impinged on the line of cutting. 
 The principle underlying this method consists of taking advantage 
 of the fact that, when heated, iron and steel can be oxidised very 
 quickly by a jet of oxygen, which jet, delivered at high pressure, 
 blows away the oxide that is made, leaving a narrow, clear cut, 
 almost as clean as a saw cut. 
 
CHAPTER II 
 LEAD BURNING 
 
 This process is one of fusion of lead by oxy hydrogen, air hycrogen, 
 or coal-gas hydrogen. The air-hydrogen process, dating back over 
 one hundred years, for a considerable time was only employed in 
 chemical work on the construction of acid tanks and vessels with 
 their pipe connections. The tanks being built of wood, lined with 
 sheet lead, the junction of the sheet was effected by this method. 
 At a later period this system was employed in large gasworks, 
 more recently for the manufacture of electric batteries and their 
 plate connections. If solder had been used on either class of work, 
 it would have been attacked by the acid. It was restricted on 
 account of its cost. 
 
 In the latter part of the year 1888 a series of experiments was 
 carried out by the Brin's (now British) Oxygen Company, with the 
 object of using coal-gas from the ordinary town main and oxygen 
 from a trade cylinder. After due consideration of the matter, it 
 was thought possible to utilise the pressure of the oxygen cylinder 
 to obtain an injection action by which the supply of coal-gas at low 
 or main pressure could be increased and thoroughly mixed with the 
 oxygen in the blowpipe previous to ignition. In order to complete 
 combustion (thus preventing the formation of carbon deposits on 
 the melted lead), after various trials had been made, the injector 
 blowpipe which we illustrate on p. 7 was designed, and was found 
 eminently adapted to the purpose. It fulfilled in other respects 
 all the requirements necessary to ensure good work, and, moreover, 
 as it produced a flame at much higher temperature than could be 
 secured by the old hydrogen-and-air system, it enabled the plumber 
 to execute about double the amount of work, of better quality, and 
 without assistance. 
 
 H (Fig. 2) is the inlet for the coal-gas supply ; shows the injector 
 inlet for the oxygen supply, fixed in position with the injector 
 removed, showing the coned end which projects through the chamber 
 into a cone formed in the body. The pipe leading away from the body 
 is screwed at the end to take the nipple, which is of various sizes, to 
 
 6 
 
LEAD BURNING 7 
 
 deal with lead from the thinnest section up to J inch thick. The 
 essential features of this type of blowpipe are the shape and design 
 of the injector cones, with their relative position to each other; the 
 reservoir for the coal supply provided between the two gas inlets 
 
 Fig. 2. — Injector Blowpipe. 
 
 and H; together with the mixing chamber, into which the brass pipe 
 is screwed. The flame also differs somewhat from that of the old 
 hydrogen type of blowpipe, its chief characteristic being a well- 
 defined pale blue cone about a quarter of an inch long. The hottest 
 
 Fig. 3. — Shoulder Taps eor Use with Lead-Burning Blowpipe. 
 
 part is the point of the cone, which, in use, impinges on the metal 
 operated upon. Melting is clean and bright. This flame is best 
 controlled by the use of what are known as " shoulder taps," illus- 
 trated herewith. 
 
 These taps in Fig. 3, marked H and respectively, correspond 
 
8 MODERN METHODS OF WELDING 
 
 with those on the blowpipe, and are connected by two lengths of 
 rubber tubing, with the twofold object of securing flexibility and 
 lightness, the pressure being previously reduced from the other 
 end of the tap, marked H. A rubber tubing leads to the ordinary 
 coal-gas supply from the main, and the oxygen tap, marked 0, is 
 similarly connected by a rubber tube with the outlet of an 
 endurance regulator, which is illustrated herewith. 
 
 This regulator in Fig. 4 is connected by a ring nut, seen at the 
 bottom, to the oxygen cylinder. The knurled nut on the right is 
 screwed into the diaphragm plate, indented on the sleeve, by which 
 means it can be adjusted to maintain any constant outflow of oxygen 
 
 Fig. 4. — Oxygen Regulator. 
 
 from the cylinder up to 30 pounds pressure per square inch. The 
 projection immediately opposite is a safety valve, the outlet or stop 
 valve being shown on top. These illustrations constitute a com- 
 plete lead-burning outfit, with the exception of the rubber tubing 
 and the oxygen cylinder. A sample injector blowpipe was submitted 
 to the Gas Light and Coke Company for trial in their chemical 
 works at BeCkton, and gave satisfactory results. A large order 
 was consequently placed. The system has met with universal 
 approval, not only in the trades already mentioned, but also for 
 roof work and general plumbing. If lead burning is required where 
 there is no coal-gas supply, this can be obtained compressed in 
 cylinders similar to those used for oxygen ; but an endurance regula- 
 tor would have to be used on the coal-gas cylinder. In lead burn- 
 ing, in order to secure sound joints in any class of work, it is neces- 
 sary that the joints are scraped perfectly clean and bright, and also 
 the strip of lead that is used for filling in the case of a butt joint. 
 
LEAD BURNING 
 
 9 
 
 For lap joints no filling strip is necessary. In the latter case, how- 
 ever, it is equally important to scrape clean all surfaces to be welded, 
 including both the overlapping edges, otherwise it is not possible 
 to do sound work. The illustrations represent the best methods of 
 executing the various types of sheet-lead burning. A (Fig. 5) is a butt 
 
 Figs. 5 and 6. — Lead-Burned Joints. 
 
 A, butt joint; B, vertical joint; C, vertical lap joint; D, lapped joint with 
 metal added. 
 
 joint. Where added metal is required, the fine of weld should be 
 about | inch to f inch wide, and thickened up. This is a very strong 
 joint. B is a vertical joint, which is lapped, and no additional 
 metal is needed. This vertical welding is not as easy as the butt. 
 The metal when molten falls very quickly, and unless the welder 
 is sharp the molten metal runs right down instead of in the 
 
10 MODERN METHODS OF WELDING 
 
 line of the joint. Much practice is necessary for this type of 
 welding. C (Fig. 6) is a vertical lap joint, too. The same 
 methods have to be adopted as with type B. This weld is not 
 so good as type B, nor so neat. D is a lapped joint with added 
 metal, the weld being about § inch to f inch wide. This is welded 
 horizontally. 
 
 In chemical works very often the sheet lead used is 20 to 30 
 pounds, and when vertical jointing has to be done with this thick- 
 ness it is usual to employ what is known as a moulding tool, which 
 is held over the lap. The lead is melted and allowed to fall into this 
 moulding tool. As soon as it is cooled, it is pushed up a bit farther, 
 melted, and filled again, the operation being repeated until the joint 
 line is finished. This moulding tool is simply a forged tool shaped 
 half-round on the inside, where it fits against the lead joint; the 
 size is about 1 inch by |- inch semicircular, about | inch thick, with 
 a handle at the other end. Below is an illustration of one. 
 
 Fig. 7. — Moulding Tool for Heavy Vertical Welding. 
 
 Lead burning, so called, is the autogenous welding of lead with 
 a gas torch. The edges of the lead parts are fused together, using 
 no solder or other metal except pure lead to fill the joint. The term 
 " lead burning " is really a misnomer, as metal is not burnt, but 
 simply fused and flowed together. This process only is employed 
 for applying lead linings of acid tanks used in chemical plants, 
 fertiliser factories, rubber reclaiming works, electroplating shops, 
 nitroglycerine huts, and other industries. The widespread use of 
 storage-battery starting systems on motor-cars has created a demand 
 in garages and battery stations for lead-welding equipment for the 
 welding of lead connectors and terminals. 
 
 Lead welding is used to some extent in making up pipe con- 
 nections; but tank and storage work are the chief applications for 
 which lead-burning equipments are purchased. Considerable skill 
 is required of a lead burner to weld lead sheets together without 
 burning holes through or causing the lead to drop. But the trade 
 is one quickly learned by an intelligent workman. Success depends 
 on clean, thorough preparation, suitable equipment, right material 
 and right working conditions, as well as personal skill. Two forms 
 of joints are used, the butt and the single-lap seams. The latter are 
 
LEAD BURNING 11 
 
 used almost exclusively in tank burning. Butt joints are used 
 mostly in lead-pipe jointing. 
 
 When making a butt joint, the edges of the sheets or pipes should 
 be trimmed straight. They may be rasped to 75 to 80 degrees in 
 opposite ways, so that when the two sheets are laid together, a 
 V groove of 30 degrees or slightly less angle is formed. If a filler 
 is to be used, the groove is filled with molten lead, flush or slightly 
 above the surface of the sheets or pipes. A preparation of the butt 
 joint by bevelling is practised for sheet burning except on very heavy 
 lead — so thick, that is to say, that it is rarely used. All thicknesses 
 up to, say, 20 pounds to the square foot are butted with square edges 
 and generally burned together without using the filler rod. The 
 use of the latter slows down the work, on account of the necessitj^ 
 of frequently cleaning off the oxide film on the rod. A joint made 
 without the filler rod is slightly thinner than the sheet; but this is 
 rarely objectionable. When burning a butt joint with the filler 
 rod, the flame is played on the sheets until the metal is softened 
 to fusing-point, but is still too stiff to flow freely. The added metal 
 is supplied from the lead filler rod to fill the V, if prepared for a 
 groove. The filling rod is held in the left hand close to the joint, 
 so that the heat of the flame melts it and deposits molten lead into 
 the joint, alternately heating the sheets and the filler rod to the 
 fusing-point, letting a drop fall into place, then whipping the torch 
 off momentarily to allow the fused metal to cool. This practice 
 is slow and seldom necessary. 
 
 It should be borne in mind that tank linings are made of chemical 
 lead — that is, new lead, free from impurities. Ordinary sheet lead 
 used for gutters and other purposes is often made of old lead, and 
 should not be used for acid tank linings which must be autogenously 
 welded. Lap seams are always used in tank linings where a smooth 
 surface is not absolutely required. In making a lap seam the top 
 sheet should lap over the under sheet about \ inch to f inch. All 
 dirt and oxide should be removed carefully from the joint with a 
 sharp scraper. Some lead burners prepare for welding by painting 
 a strip of asphaltum along the edge about 1 inch wide, letting it 
 dry, and then scraping it off about \ inch width next to the edge with 
 a sharp scraper, which removes the oxide and the asphaltum. The 
 asphaltum tends to hold from breaking down when overheated, and 
 thus favours the use of a large flame burning at a fast rate. The 
 flame is directed against the top sheet about \ inch from the edge. 
 Some burners give the torch a circular motion, the diameter of the 
 circular path being from \ inch to | inch ; but the most rapid work 
 
12 MODERN METHODS OF WELDING 
 
 on level seams is done with no movement except forward. The 
 top sheet fuses and unites with the metal beneath, the edge breaks 
 down, and the corner is filled with molten lead, which adheres to 
 the lower sheet, making a weld about § inch wide. The burner 
 dispenses with the use of the lead filler rod or solder stick on lap 
 seams, simply melting down the edge of the overlapping sheet, 
 causing it to unite with the one beneath. 
 
 In vertical lap seams the torch is generally given an in-and-out 
 motion, thus taking the flame close to the lead and then away. 
 When the flame is close to the lead a small section of the upper sheet 
 over the seam melts, and slides down about \ inch, where it cools 
 and unites. The repetition of the torch movement causes this 
 action to be repeated, drops of molten lead sliding down the joint, 
 uniting each time the torch is held close. This method is used only 
 on vertical work, where it is not feasible to do straightaway burn- 
 ing. Some welders give the circular motion instead of the in-and-out, 
 but the result is practically the same. 
 
 The temperature needed for lead burning is low in comparison 
 with that required for welding cast iron or steel, being only 62° F. 
 A special torch is employed, which is held in the hand and manipu- 
 lated somewhat the same as a soldering iron. These torches may 
 be used with acetylene successfully. 
 
 The lead burner necessarily becomes an expert at working lead 
 within narrow temperature ranges. He must heat the metal at the 
 joint to the fusing-point, but not so much that it becomes liquid 
 and drops away. When butt welding with the solder stick, he must 
 avoid heating and adding metal so hot that when it drops into the 
 joint it burns a hole through instead of filling the groove. He should 
 make provision for growth or expansion of the finings of the hot 
 tanks. Repeated heating and cooling cause the lead to expand. 
 After a period of use the lining will be found in a corrugated condi- 
 tion, due to permanent expansion. 
 
 Lead burning of storage-battery connections is comparatively 
 easy, and is quickly mastered by anyone having some mechanical 
 skill. Handy men soon become sufficiently expert in this class of 
 work to burn storage-battery terminals successfully. The danger 
 to be avoided is overheating. The operator must learn to clean 
 thoroughly the parts to be united, and to apply the flame no longer 
 than is required to fuse the metal and bring about union. When 
 the connections have been flowed together, the excess metal should 
 be wiped off with a woollen cloth greased with mutton-fat or beef- 
 tallow. 
 
CHAPTER III 
 
 MANUFACTURE OF OXYGEN 
 
 Oxygen is invariably the principal ingredient or combustion agent 
 used in the cutting of iron and steel in autogenous welding with the 
 blowpipe. It is necessary that those handling the blowpipe be 
 familiar with its properties, manufacture, storage, and methods of 
 use. Oxygen is the most widely distributed of all bodies. It 
 exists in a state of mixture in the air, which contains one-fifth of its 
 volume of this gas. Water is a compound of oxygen and hydrogen. 
 It is a colourless, tasteless, and odourless gas. One litre of oxygen 
 at 0° C. and atmospheric pressure weighs 1 43 grammes, its density 
 is 1-1056, its chemical symbol 0, its atomic weight 16. The 
 characteristic property of oxygen is its power of supporting com- 
 bustion ; a glowing candle will instantly burst into name if plunged 
 in a jar of oxygen. 
 
 Iron heated to redness burns readily in air or oxygen. This 
 unique phenomena is the whole secret of the cutting of iron and steel 
 with a jet of oxygen (which will be described under " Cutting Iron 
 and Steel " in a later chapter). The combustion is a chemical 
 reaction between the oxygen and the body which burns with it. 
 The product of the combustion is called " oxide." 
 
 The manufacture of oxygen can be carried out by several pro- 
 cesses, some of which are the barium oxide, the electrolysis of water, 
 and the Linde process of the fractional distillation of liquid air, 
 which brings about the production of pure oxygen. In 1886, on 
 the present site of the British Oxygen Companj^'s works in West- 
 minster, a large oxygen plant was erected, which must take a lead- 
 ing position in any article dealing with industrial gases. The plant 
 was installed by Brin's (now the British) Oxygen Company. 
 Although in itself a qualified success, it led to the development by 
 that Company of a process which was destined to supersede all other 
 known methods of manufacturing oxygen, and greatly to enhance 
 the commercial value of the gas. 
 
 The Brin process of producing oxygen is based on the barium- 
 oxide process, first suggested by the eminent chemist Boussingault 
 in 1857. Boussingault discovered that at a temperature of about 
 
 13 
 
14 MODERN METHODS OF WELDING 
 
 1,000° F. the monoxide of the metal barium would absorb oxygen 
 readily from the atmosphere, with the resulting formation of the 
 dioxide, and that at a higher temperature of about 1,600° E. the 
 oxygen thus absorbed would be given off again, and monoxide 
 restored to its original condition, ready for the cycle to be repeated. 
 Continuous efforts were made to establish a commercial process for 
 the production of oxygen on this apparently indestructible reaction 
 of barium oxide. In spite, however, of its chemical simplicity, 
 many practical difficulties arose, which remained unsurmounted 
 until the event of the Brin's Oxygen Company in 1886. The Com- 
 pany was formed to work the patents of the two French chemists 
 whose names it bore. Initial experiments, conducted on a small 
 scale under these patents, had met with considerable success, so that 
 a large plant was laid. 
 
 It will, however, be seen from what has already been stated that 
 the Company's old title was always somewhat of a misnomer. The 
 process which has made them not only the oldest and leading gas 
 compressors of the day, but also the pioneers of the gas cylinder in- 
 dustry, has always been more British than Brin. Although now only 
 of historical interest, the barium process is entitled to more than a 
 passing recognition in any description of the development of industrial 
 oxygen, because the industry was not only founded, but was success- 
 fully conducted by means of that process for more than twenty years, 
 during the whole of which time, although many other methods of 
 producing oxygen were proposed and tried in this country and 
 abroad, the barium process remained in sole possession of the field. 
 Furthermore, it is worthy of note, in these days when it is customary 
 to credit any country but our own with industrial enterprise, that 
 in Germany, France, and the United States, the oxygen industry 
 was started with the barium, designed and erected by the British 
 Oxygen Company, who have acquired the British patents of Pro- 
 fessor Linde and Dr. Hampson. These two inventors are the authors 
 of the self -intensive system of liquefying air, on which are based the 
 numerous processes introduced in recent years, with extravagant 
 and preposterous claims that have gone far to bring liquid air 
 into disrepute. A serious scientist saw long ago the only really 
 valuable commercial application of oxygen and nitrogen. For 
 many years he devoted himself to adopting his system of liquefying 
 air for this purpose. That his labours have been crowned with 
 success has been proved by the fact that on the Continent the com- 
 pany which controls his patents has already erected a large number 
 of plants, which are in daily operation, giving most satisfactory 
 
MANUFACTURE OF OXYGEN 15 
 
 results. It is the Linde plant that the British Oxygen Company 
 erected at Westminster and many other great centres. 
 
 Briefly expressed, the process consists of liquefying the air com- 
 pletely in the first instance by the self-intensive system. Whilst 
 obtaining almost complete transference of heat from the compressed 
 air entering the apparatus to the liquefied air, the liquid is submitted 
 to a special process of rectification, by means of which oxygen of 
 any degree of purity up to 98 or 99 per cent, can be obtained. The 
 plant is driven by a Diesel engine, which develops som 35 h.p. 
 Air is compressed by three stage compressors, belt-driven. Between 
 each stage of compression the air is restored to normal temperature 
 by passing through coils contained in a cooler, through which water 
 circulates. The system of purification of the air is very complete, 
 all the moisture and carbolic acid being practically eliminated, 
 first in the purifiers containing unslaked lime and calcium chloride, 
 while the final traces of moisture are subsequently removed by 
 freezing in a forecooler, which is employed, partly for this purpose, 
 partly to precool the air before it enters the separators. The lower 
 part of the forecooler is reduced in temperature by a small C0 2 
 refrigerating machine of the usual type made by the Linde British 
 Refrigerating Company, but the upper part is cooled by the waste 
 nitrogen from the separator. 
 
 The interchange is so effective that, before the nitrogen is 
 returned to the atmosphere, it has taken so much, heat from the in- 
 coming compressed air that it leaves the forecooler only a few degrees 
 below normal temperature. The compressed air, on the other hand, 
 leaves the bottom of the forecooler at a temperature considerably 
 below freezing-point, and then enters the top of the separator, passing 
 downwards to a series of coils, which are so constructed as to be 
 surrounded by both the outgoing cold gases. The bottom of the 
 separator contains liquid air, or, more correctly speaking, liquid 
 oxygen. The compressed air, on its way to the expansion point, is 
 conveyed through the liquid, by which means it is largely con- 
 densed. It then passes through the regulating valve, at which 
 point it expands, and is ultimately discharged into the top of an 
 inner central chamber which forms the rectifying column, in which 
 the separation of oxygen and nitrogen is effected, oxygen descend- 
 ing in a liquid state to the bottom of the separator, nitrogen ascend- 
 ing in a gaseous or vaporous condition to the top. 
 
 The nitrogen is allowed freely to discharge into the atmo- 
 sphere through the forecooler, as already explained; the oxygen is 
 allowed to boil off in any desired quantity by the adjustment of a 
 
16 MODERN METHODS OF WELDING 
 
 discharge valve. Both gases, however, on leaving the separator, 
 are kept in intimate contact with the cells conv^ing the incoming 
 air, so that before leaving the apparatus the heat of the incoming 
 air has been mostly transferred to the gases. The plant is very 
 conveniently arranged, and consists of two separators and two fore- 
 coolers, one of each being worked at a time. In this way continuous 
 working is ensured, for when ice (due to entrapped traces of moisture 
 in the air) has accumulated to such an extent as to cause a stoppage 
 in one separator, the other can be put in operation, whilst the first 
 is allowed to thaw. In practice, freezing is found to occur after six 
 to seven days of continuous work. 
 
 The description here given represents the normal working of the 
 plants. But before pure oxygen can be produced the separator 
 has to be cooled down, and a considerable quantity of liquid pro 
 duced. This operation takes about three hours, during which time 
 the compressed air circulating through the coils is (by the adjustment 
 of the regulating valve) maintained at a pressure of about 2,500 
 pounds per square inch. Afterwards, when the oxygen is being 
 produced, the pressure is about 800 pounds per square inch, at which 
 pressure it continues to work. 
 
 Oxygen obtained from liquid air may contain more or less nitro- 
 gen. That obtained by the electrolysis of water might contain a 
 little hydrogen. These two gases are considered as impurities. 
 If hydrogen were present to an appreciable extent, it would have the 
 disadvantage of forming with the oxygen an' explosive mixture. 
 It has been demonstrated that in the cutting of iron and steel by the 
 blowpipe cutters, the presence of nitrogen, even in small quantities, 
 has an adverse effect on the quality and rapidity. Oxygen com- 
 pressed in cylinders is generally delivered containing 96 to 99 per 
 cent, of oxygen; but the commercial guarantee may be as low as 
 95 per cent. The British Oxygen Company guarantee the quality 
 of their oxygen as 98-5 to 99 per cent., and many experiments carried 
 out by the author confirm this. 
 
 Analysis is conducted by acting on a definite volume of gas 
 with a chemical which rapidly absorbs the oxygen and leaves the 
 impurities intact (hydrogen, nitrogen, or carbon dioxide). The 
 quantity of gas absorbed, compared with the original volume, gives 
 the degrees of purity. The analysis is generally done in graduated 
 test-tubes, as a rule of 100 centimetres capacity and graduated 
 to 100. The absorbent liquid takes the place of the oxygen ab- 
 sorbed, so that when the reaction is over it is only necessary to read 
 off the level of the liquid to know the purity of the oxygen. 
 
CHAPTER IV 
 MANUFACTURE OF HYDROGEN 
 
 These two gases — -oxygen and hydrogen — are now manufactured 
 on very large scales. Hydrogen can be very much more freely 
 obtained than in the past, and will be more used in future for weld- 
 ing purposes. There are very few who are acquainted with oxy- 
 hydrogen as applied to welding at the present time, although it was 
 in extensive use until the advent of calcium carbide in large com- 
 mercial quantities, which generate the acetylene gas. The latter 
 was employed, in combination with oxygen, for use in the blow- 
 pipe for welding, giving a very much higher temperature of 6,000° F. 
 against the oxy-hydrogen flame's 4,000° F. 
 
 There are many advantages in using oxy-hydrogen blowpipes. 
 Firstly, the hydrogen is supplied in cylinders, the same as oxygen. 
 Therefore it can practically be used anywhere. There is no wasted 
 gas, no expense in initial outlay for generators or fixing of piping, 
 no messy substance to clear away, as in acetylene generators, the 
 two gases, oxygen and hydrogen, under equal pressures, thus ensur- 
 ing a constant, steady flame. This blowpipe is also well adapted for 
 the burning of lead and the welding of aluminium, because the heat 
 of temperature is not so high, but is suitable for these metals. How- 
 ever, there is the disadvantage of not being able to get high enough 
 temperature to weld heavy mild steel or cast iron, as the heat is 
 absorbed and there is loss by conductivity. It is practically im- 
 possible to weld steel plates exceeding T V inch thick. 
 
 The flame is produced by the mixing of two volumes of hydrogen 
 and one of oxygen. This is the theoretical mixture; but the author 
 found in practice that it required, to maintain a good heat to com- 
 plete a proper weld, three parts of hydrogen to one of oxygen. 
 Hydrogen and oxygen have strong affinity for each other. Their 
 combination occurs with great explosive violence, the product being 
 water in the form of steam. This steam is, of course, superheated 
 in the hot flame, which process of dissociation involves a consumption 
 of heat, which is abstracted from the flame, hence an oxidising effect 
 on the weld. This can only be avoided by feeding the flame with 
 
 17 2 
 
18 MODERN METHODS OF WELDING 
 
 an excess of hydrogen (in practice four to five volumes of hydrogen 
 to one of oxygen), whilst, on the other hand, only that amount of 
 heat can be obtained which is disengaged in the combination of the 
 two volumes of hydrogen and one of oxygen. The large excess of 
 hydrogen required to prevent oxidation furnishes in itself one way 
 why welding by means of the hydrogen-oxygen flame is frequently 
 uneconomical. A further reason, however, lies in the fact that the 
 temperature of the flame, in consequence of this excess of hydrogen, 
 is essentially lower than it ought to be. In spite of this, it is found 
 in practice that for thin plate welding the oxy-hydrogen blowpipe 
 has certain advantages. The flame is more diffused than in the case 
 of oxy-acetylene flame, hence less liable to melt through and pierce 
 the metal. 
 
 Every student, if he is to become a welder, must have a fair 
 knowledge of all the materials used and their constituents, as also 
 of the gases such as oxygen, hydrogen, and acetylene. Since these 
 gases are usually provided in convenient cylinders for commercial 
 purposes, the users as a rule are not acquainted with their manu- 
 facture or with the source of supply. But it is really essential that 
 the users of these gases should be acquainted with the process of 
 manufacturing them. It is probable that the war, which has brought 
 home to many engineers and business men how easily hydrogen and 
 oxygen can be generated, will lead to an increased demand for. these 
 gases and to their wider appreciation in industrial processes. 
 
 Where both the decomposition products of water, hydrogen, 
 and oxygen can be utilised, electrolytic generation offers advantages 
 over the isolation of hydrogen from water-gas, which is a cheaper 
 method when worked on a really large scale. In the laboratory 
 the electrolysis of water is a very simple matter. The electrodes 
 are placed in acidulated water in a glass cell. Each electrode is 
 surrounded by a hood (an inverted test-tube, e.g., in which the gas 
 evolved collects). Currents of about two volts give one volume of 
 oxygen and two of hydrogen. 
 
 The technical electrolyser is not quite so simple as the laboratory 
 cell. The design and construction of automatic apparatus which 
 will continuously yield both gases in a pure condition of 90 per cent, 
 and more, practically without requiring any attendance, have left 
 sufficient scope for the ingenuity of inventors. Glass cells being 
 out of the question, the container has to be adapted to the electro- 
 lyte. Water itself is a poor conductor to serve as an electrolyte. 
 Either sulphur acid or caustic, both of about 20 per cent, concen- 
 tration, are used; but the chemicals are merely to increase the con- 
 
MANUFACTURE OF HYDROGEN 19 
 
 ductivity of the water, which is the electrolyte, and has to be re- 
 plenished by feeding distilled water into the cells. The impurities 
 of ordinary water, notably chlorine, would tend to corrode the 
 apparatus. 
 
 The acid is placed in lead containers, the alkali in iron cells. 
 The reversible decomposition of the water itself would only require 
 1 -23 volts ; but the potential applied has to overcome the resistance 
 of the leads and of the electrolyte, and further the polarisation of 
 the electrodes by the gases liberated. Sulphuric acid is a better 
 conductor than alkali, but the supertension of the lead cathode is 
 decidedly higher. Hence, on the whole, the caustic alkali requires 
 a lower decomposition potential. As regards gas purity and, to a 
 certain extent, the efficiency, there is not much to choose between 
 the acid and the alkali processes, though the number of kilowatt 
 hours required for the liberation of 35 cubic feet of the two gases 
 varies from 6 down to about 3-5, at volts ranging from 3-6 down 
 to 2-3. But in all these figures the decimals become important, 
 and there are features justifying a careful selection of a type of 
 electrolyser. 
 
 The general preference is for alkali cells. In the Schoop acid 
 electrolysers, the two groups of electrodes, anodes, and twice as 
 many cathodes, are all lead tubes, open below to let the acid act 
 on the lead wires in the tube. Each tube is insulated by a refractory 
 hood. The groups are coupled in parallel. 
 
 In the Garute alkali electrolyser, vertical diaphragms of iron 
 separate the iron tank into anode and cathode compartments, 
 the two groups of iron, electrodes being again in parallel. 
 
 The Schukert electrolyser uses insulating diaphragms and places 
 iron hoods in the different compartments to collect the gases. 
 
 The Schmidt cell of the Oerlikon Company is of the filter-press 
 type, and the corrugated iron electrodes are coupled in series. They 
 are, in fact, bipolar, as in some electrolytic copper baths and in 
 bleach cells. 
 
 The electrolysers of the Integral Oxygen Company, London, 
 illustrated on p. 20, differ from types mentioned above in so far 
 as each cell is self-contained, in having two electrode units, and, 
 further, as to arrangements made for the regulation of the gas 
 pressure. This is an Integral unit generator. The cells are of the 
 Hayard-Flamand type. The early patents of this cell date back to 
 1900, but the particulars, the feed and pressure regulation, are later 
 inventions. The battery photograph seems to suggest a filter-press 
 arrangement, but each cell is independent, and the common features 
 
20 
 
 MODERN METHODS OF WELDING 
 
 are the water and the gas pipes and the grouping in series. Each 
 cell is independent, each has its own feed and discharge devices, 
 each consists of a thin, rectangular frame of cast iron, to which two 
 cast-iron plates, the electrodes, are bolted, mica insulators being 
 interposed. 
 
 The asbestos diaphragm divides the sj>ace of the flat cell vertically 
 into two halves, an anode compartment and a cathode compartment. 
 Each compartment communicates through Wo ports with a gas 
 chamber. The right is the water chamber. Several jars and pipes 
 
 Fig. 8. — 4, 000 -Unit Geneeatoes eoe Pboducing Oxygen and Hydeogen. 
 
 will be seen on the top of the gas chamber. The gas leaves through 
 the two outer glass jars — the hydrogen through the jar on the right, 
 the oxygen through that on the left — and enters the two gas pipes 
 which are in, but not on, the sketch. A third pipe is the distilled- 
 water pipe. This is connected with the central jar through which 
 the cell is originally charged with caustic soda of 20 per cent. The cell 
 discharges a spray of gas and caustic soda ; this frothy liquid spreads 
 upon the base plate, which has a raised edge, this ridge preventing 
 the liquid from running back into the port and choking it. It 
 spreads over to the port and there flows back into the cell. Thus 
 a kind of circulation is set up ; the gas itself is further deprived of its 
 
MANUFACTURE OF HYDROGEN 21 
 
 moisture by having to force its way through a gas tap in the glass jar, 
 which contains an inverted bell of iron. The liquid is intercepted 
 between this bell and the jar, the top of which is joined to the gas 
 pipe. The cells are worked at a temperature of about 55° C. The 
 guaranteed purity of the gases evolved is oxygen 99 per cent., 
 hydrogen 99-5 per cent. During recent trials at Farnborough the 
 oxygen purity was maintained at 99-8 per cent., which was an 
 exceptionally high figure. 
 
 It must be noted that the gases are not purified in any way, but 
 enter the gas holders or cylinders as they leave the glass jars. The 
 average potential is about 2-2 or 2-3 volts per cell, at the normal 
 current of about 600 amperes. Each cell has a rated output of 
 4-8 cubic feet of oxygen and 9-6 of hydrogen, and gives an average 
 yield of 4 cubic feet of oxygen, and twice as much hydrogen. 
 
CHAPTER V 
 MANUFACTURE OF CARBIDE 
 
 The manufacture of calcium carbide is carried out on such exten- 
 sive lines that any student of welding should make himself conver- 
 sant with its manufacture. Acetylene was discovered by Davy 
 in the chemical laboratory in the year 1836. Many methods of 
 preparing the gas were described by various experimenters, the 
 majority of which are only of academic interest. But it may be 
 mentioned that in 1840 Hare, by heating in an electric arc a black 
 residue obtained by heating a mixture of mercuric cyanide and 
 lime, arrived at a compound which was undoubtedly calcium car- 
 bide, although not recognised as such at the time. 
 
 The next date of importance is the year 1862, in which the German 
 chemist Woehler (whose name is well known amongst chemists as 
 the discoverer of synthetic urea and other important bodies, includ- 
 ing aluminium) obtained " calcium of carbide " decomposed water 
 with formation of calcium hydrate and acetylene. It is hard, 
 really, to say who was the actual discoverer of calcium carbide. 
 It was Moisson, " the king of experimental savants," who published 
 his classic investigations. His results were obtained as factors in a 
 magnificent research, every step in which was logically worked out 
 and verified; a research which will ever stand out as a scientific 
 classic. But the fact remains that he only attained and published the 
 discovery of the direct formation of calcium carbide in the electric 
 furnace to find that his work had been forestalled by a few months 
 by the chance observation of an engineer who, although devoid of 
 chemical knowledge, yet had sufficient acumen to grasp the commer- 
 cial importance of the discovery. Anyone who, with a mind free 
 from prejudice, reads the evidence on this subject, is forced to the 
 conclusion that the world owes " commercial acetylene " to the 
 Canadian engineer, Willson, and the shrewd business men who 
 supported him. 
 
 Moisson obtained crystallised calcium carbide during a systema- 
 tic and masterly research upon the products of the electric furnace. 
 His first paper described an electric furnace, which he distinctly 
 stated was not an industrial apparatus, but for research purposes 
 
 22 
 
MANUFACTURE OF CARBIDE 23 
 
 only. His work consisted in the preparation of crystallised metallic 
 oxides such as those of calcium, strontium, barium, magnesium, 
 aluminium, iron, chromium, etc. He then dealt with fusion and 
 volatilisation of some refractory bodies and metals. This was 
 followed by his classical research on the different varieties of carbon 
 and the formation of the diamond. The description of the forma- 
 tion of calcium carbide is included in a study of the carbides, 
 silicides, and the borides. After a brief review dealing with the 
 work of Berthelot, Woehler, and Travers, who in 1893 obtained a 
 mixture containing some carbide, Moisson writes : 
 
 " The question had reached this point when, in a note appearing 
 in the Comptes Rendus cle P Academie des Sciences on December 12, 
 1892, I made public for the first time the formation in the electric 
 furnace of a carbide of calcium fusible at a high temperature. 
 This is what I wrote on the subject. If the temperature reaches 
 3,000° C, even the furnace material, the quicklime, melts and flows 
 like water. At this temperature the carbon quickly reduces the 
 oxide of calcium (lime), the metal itself is liberated in abundance; 
 it unites readily with the carbon of the electrodes to form a carbide 
 of calcium, liquid at a red heat, which is easy to recover. Follow- 
 ing this research, I presented to the Academie des Sciences a note 
 of carbide of calcium on March 5, 1894, another note of carbides 
 of barium and strontium on March 9, 1894. In this work I showed 
 that in a high temperature of the electric furnace there exists only 
 one compound of carbon and calcium. This compound was crystal- 
 line; I established its formula by analysis, and in the study of its 
 properties I made it clear that the bodies decompose in water, 
 liberating the gas acetylene absolutely pure." 
 
 This was the starting-point of the acetylene industry. 
 
 At the end of a patent, No. 492,377, U.S.A., on the preparation 
 of aluminium bronze, M. Thos. Willson made an allusion to an inter- 
 minate carbide of calcium as well as to a great number of other 
 bodies, elements or compounds, but did not give an analysis of the 
 two compounds obtained, nor even mentioned that this product 
 decomposes cold water with liberation of any gas whatever. He 
 also avoided, with the greatest care, that " bath of fusion " brought 
 metallic calcium. 
 
 Moisson prepared his carbide by making an intimate mixture of 
 120 grammes of lime from marble, and 70 grammes of carbonised 
 sugar, heating for fifteen minutes in a crucible in the electric furnace 
 with a current of 350 amperes and 70 volts. He used a slight excess 
 of lime to counteract the carbon obtained from the crucible. He 
 
24 MODERN METHODS OF WELDING 
 
 noted that if any impure lime was used, containing sulphates, phos- 
 phates, or silica, these impurities would be found in the acetylene 
 liberated from the carbide. Moisson carried out a number of experi- 
 ments with impure materials and analysed the gas obtained; also 
 studied the physical qualities of the carbide, the gas yield as well 
 as its chemical properties and that of acetylene. It is interesting 
 to note that he states that " pure and dry " nitrogen does not react 
 with carbide even at 1,200° C. Had he continued this experiment 
 to a higher temperature he would have probably been the discoverer 
 of cyanamide. 
 
 Whatever the facts of the case may be, Moisson's systematic 
 and scientific research work and the practical results obtained 
 place his name for ever in an unassailable position in the history of 
 the industry. From the time when Moisson and Willson published 
 their investigations, the accepted equation for the production of 
 calcium of carbide in the electric furnace from carbonaceous matter 
 and lime has always been — 
 
 CaO+3C = CaC 2 +CO. 
 
 This final result may be approximately the production of two 
 compounds, CaC 2 and CO; but the reaction, or rather, reactions 
 which take place before the final stage is reached certainly appear 
 to be far more complex. The conversion of the raw material into 
 carbide appears to take place in steps. 
 
 In the past, every person who had cheap water-power available 
 believed that he had the Alpha and Omega for making cheap carbide, 
 even if there was no limestone or carbon within hundreds of miles. 
 To equip a modern factory, a very large capital is required, and a 
 suitable site, where cheap water-power is available, near to lime- 
 stone, coal, and coke. The power for electric energy, in the case of 
 every carbide factory in existence, is supplied from a hydro-electric 
 station, where the prime mover, or force, is falling water — in other 
 words, a waterfall. This falling water passes through turbines or 
 Pelton wheels. The type of turbine is decided by the head or volume 
 of water available. These turbines drive dynamos or electric 
 generators ; some of these generators are used in the electric furnaces 
 for the melting of the carbon and the lime for producing the carbide 
 of calcium. 
 
 An electric carbide furnace as a rule consists simply of a steel case 
 or box, or a steel-framed case or box lined inside with suitable 
 refractory material, having a tapping hole very similar to a tapping 
 hole in an iron-furnace or cupola. In some of the furnaces electrodes 
 are fixed in the bottom of the furnace; in others the electrodes are 
 
MANUFACTURE OF CARBIDE 25 
 
 suspended from the top. Each of the electrodes is held in various 
 kinds of holders, which are connected to busbars. The electrodes 
 are made of carbon. The materials used in the manufacture of 
 carbide of calcium are primarily — ■ 
 
 (a) Carbon for the charge in the form of coal or coke, 
 
 (b) Carbon for the electrodes, 
 
 (c) Lime. 
 
 Coal. — Only one form of coal, anthracite, has up to the present 
 given anything like satisfactory results. An anthracite coal con- 
 taining not more than 4 per cent, of ash and not more than 0-040 
 per cent, of phosphorus will serve admirably in a modern furnace. 
 
 Limestone. — Mountain limestone is the stone composing the 
 limestone ranges known to most of us. The best stone is found in 
 the Buxton and Settle Hills districts. There is also oolitic limestone 
 (so called because of the resemblance to a mass of fish roe), made up 
 of small rounded grains of carbonate of lime. Chalk is fine-grained 
 limestone, consisting of finely commuted shells. Magnesia lime- 
 stone is a limestone mixed with more or less magnesia. This is 
 very suitable for carbide-making. The principal impurities present 
 in limestone used for carbide-making are magnesia, alumina, silica, 
 iron oxide, sulphur, and alkalies. The maximum quantities of 
 impurities permissible in good limestone for carbide should be 
 about 0-50 per cent, of magnesia (MgO), 0-50 per cent, of alumina 
 and iron oxide, 0-01 per cent, of phosphoric acid (P 2 5 ), about 1 to 
 1-2 per cent, of silica (Si0 2 ), and only traces of sulphur. This 
 corresponds to a stone containing about 97 to 98 per cent, of 
 carbonate of lime. 
 
 Carbide, as now made by a first-class factory, yields 310 litres 
 per kilogramme (4-95 feet per pound), and contains, therefore, 
 89 per cent, of pure carbide, 11 per cent, of impurities. A metric 
 ton of commercial carbide will contain 890 kilogrammes of pure 
 carbide. To produce this 890 kilogrammes of pure carbide in a 
 metric ton, based on the lime of 96 per cent, purity, coke with 6 per 
 cent, of ash, the following will be theoretically the quantities of raw 
 materials for a metric ton of carbide — 
 
 56 890 
 Lime 7T7 x 77-777, =811 kilogrammes. 
 64 0-96 ° 
 
 ni 36 890_. 
 Coke 64 0^96 ~ 521 
 Total kilogrammes 1,332 =2,797 pounds. 
 
 (1 kilogramme =2-2 pounds.) 
 
26 MODERN METHODS OF WELDING 
 
 Therefore it takes 1,332 kilogrammes of materials to produce 
 890 kilogrammes of commercial casbide. Carbide is produced in 
 the electric furnace by the fusion of coal approximate!} 7 30 parts 
 by weight, lime 50 parts by weight, to a temperature reaching 
 4,000° C. 
 
 Under the enormous temperature of the electric furnace (which 
 is carried by the carbon electrodes) the lime and carbon combine. 
 The liquid carbide which results is tapped from the furnace (the same 
 as the blast furnace or foundry cupola) into receptacles which, 
 when cooled, are transported to the crushing machines, which break 
 them up. Then on the mechanical graders, which separate the 
 various sizes; the carbide is then packed into drums, and the 
 covers are hermetically sealed. 
 
 The formula of carbide of calcium is represented by CaC 2 ; it 
 is made of 62-5 per cent, of calcium. 
 
 Use 1-4 pints of water to each pound of carbide in water-to- 
 carbide generators, 5 pints of water in carbide-to-water generators. 
 This should assure full decomposition. Carbide should yield 
 4-8 cubic feet of acetylene for every pound of carbide placed in 
 the generating chambers. There are various apparatus for testing 
 the carbide on the market. When a sample is tested it must be 
 broken up in a mortar to about J-inch .mesh, then screened to 
 remove the dust as rapidly as possible, then weighed out in a limited 
 quantity accurately by a chemist's scale. The carbide is then 
 placed in the generating chambers immediately, the water turned 
 on (the height of the bell should be marked before the water 
 is turned on); then, when the generation has ceased, the bell 
 should be marked and then measured with the already provided 
 instruments. 
 
CHAPTER VI 
 ACETYLENE 
 
 Acetylene or, as it is scientifically named, "ethine," is a simple 
 hydrocarbon consisting of 24 parts by weight of carbon and 
 2 parts by weight of hydrogen; its chemical symbol, C 2 H 2 , 
 meaning it is a compound of two atoms of carbon combined with two 
 atoms of hydrogen. It is a clear, colourless gas, of a specific gravity 
 of 0-92. It is, owing to its synthetic formation, the most pure, 
 at the same time nearly the richest hydrocarbon gas, containing 
 no less than 92-5 per cent, of carbon when perfectly pure and free 
 from water vapour. It has an illuminating value of 50 candle- 
 power per cubic foot. 
 
 It has a most unmistakable and penetrating odour. When 
 present in the proportion of only 1 part in 10,000 parts of air it 
 is distinctly perceptible long before there is sufficient gas present 
 to cause explosion. Therefore it can be at once detected, and an 
 explosion prevented. One burner passing 1 cubic foot per hour 
 in a room of 2,500 cubic feet area for a period of nine or ten hours 
 would not be sufficient, with the same quantity of air, to make an 
 explosive mixture. The largest acetylene burners only pass 1 cubic 
 foot per hour, against the 5 cubic feet of coal-gas. It is, therefore, 
 obvious that the prevailing popular belief as to acetylene being 
 more dangerous than coal-gas is a fallacy. 
 
 Acetylene and oxygen ignite at a temperature of 400° C. The 
 temperature of combustion is 4,000° C. Acetylene, although 
 practically pure gas, usually contains some impurities in a greater 
 or less proportion, mostly sulphuretted and phosphuretted hydrogen 
 due to the presence of sulphate of calcium, gypsum, and calcium 
 phosphide in the lime, or to the sulphur and phosphorus in the 
 coal and coke. Acetylene is always contaminated with ammonia, 
 formed by the combination of nitrogen derived from the coke with 
 the hydrogen of the water during decomposition of the carbide. 
 That acetylene is a poisonous gas has been proved to be untrue. 
 When pure it is relatively harmless. The range of explosibilitj' 
 is wider in the case of acetylene than of coal-gas. Mixtures having 
 less than 5 and more than 60 per cent, are practically non-explosive. 
 
 27 
 
28 MODERN METHODS OF WELDING 
 
 Acetylene, being an endothermic compound, is liable, when pure, 
 if compressed without at the same time being cooled, to explode 
 spontaneously. Acetylene is soluble in water and many other 
 liquids. It can be liquefied at a pressure of about 325 pounds per 
 square inch and forms a mobile and highly refractory liquid, much 
 lighter than water. 
 
 The reactions that occur when carbide and water are brought 
 into contact belong to the class which chemists usually term double 
 decompositions. Calcium carbide is a chemical compound of a metal 
 calcium with carbon containing one chemical " part," or atomic 
 weight, of the former united to two chemical parts of the latter. 
 Its composition is expressed in symbols by CaC 2 . Similarly, water 
 is a compound of two chemical parts of hydrogen with one of oxygen, 
 its formula being H 2 0. When these two substances are mixed to- 
 gether, the carbon of the calcium carbide leaves the calcium unit- 
 ing together to form that particular compound of hydrogen and 
 carbon, or hydrocarbon, which is known as acetylene, whose formula 
 is C 2 H 2 , while the residual calcium and oxygen join together to 
 produce calcium oxide of lime (CaO). Put into the usual form of an 
 equation, the reaction proceeds thus : 
 
 CaC 2 +H 2 0=C 2 H 2 +CaO. (1) 
 
 This equation not only means that calcium carbide and water 
 combine to yield acetylene and lime; it also means that one chemical 
 part of carbide reacts with one chemical part of water to produce one 
 chemical part of acetylene, one of lime. But these four chemical 
 parts, or molecules, which are equal chemically, are not equal in 
 weight, although, according to the common law of chemistry, they 
 each are a fixed proportion one to the other. Hitherto, for the sake 
 of simplicity, the by-product in the preparation of acetylene has been 
 described as calcium oxide, or quicklime. It is, however, one of the 
 characteristics of this body to be hygroscopic, or greedy of moisture, 
 so that if it is brought into the presence of water, either in the form 
 of liquid or vapour, it immediately combines therewith to yield 
 calcium hydroxide, or slaked lime, whose chemical formula is 
 Ca(OH) 2 . Accordingly, in actual practice, when calcium is mixed 
 with an excess of water, a secondary reaction takes place over and 
 above that indicated by equation (1), the quicklime produced com- 
 bining with one chemical part or molecule of water, thus : 
 
 CaO+H 2 = Ca(OH) 2 . 
 
 As these two actions occur simultaneously, it is more usual and 
 more in agreement with the phenomena of an acetylene generator 
 
ACETYLENE 29 
 
 to represent the decomposition of calcium carbide by the combined 
 equation : 
 
 CaC 2 +2H 2 0=C 2 H a + Ca(HO) 2 . (2) 
 
 By the aid of calculations analogous to those employed in the 
 preceding paragraph, it will be noticed that equation (2) states that 
 1 molecule of calcium carbide, or 64 parts by weight, combines 
 with 2 molecules of water, or 36 parts by weight, to yield 1 molecule, 
 or 26 parts by weight, of acetylene, and 1 molecule, or 74 parts by 
 weight, of calcium hydroxide (slaked lime). Here, again, if more 
 than 36 parts of water are taken for every 64 parts of calcium carbide, 
 the excess of water over these 36 parts is left undecomposed; and 
 in the same fashion, if less than 36 parts of water are taken for every 
 64 parts of calcium carbide, some of the latter must remain un- 
 attacked, whilst, obviously, the amount of acetylene liberated cannot 
 exceed that which corresponds with the quantity of substance suffer- 
 ing complete decomposition. If, for example, the quantity of water 
 present in a generator is more than chemically sufficient to attack 
 all the carbide added, however large or small that excess may be, 
 no more, and, theoretically speaking, no less, acetylene can ever 
 be evolved than 26 parts by weight of gas for every 64 parts by 
 weight of calcium carbide consumed. It is, however, not correct to 
 invert the proposition, and to say that if the carbide is in excess of 
 water added, no more, and, theoretically speaking, no less, acetylene 
 can be involved than 26 parts by weight of gas for every 36 parts 
 of water consumed, as might be gathered from equation (2) ; because 
 equation (1) shows that 26 parts of acetylene may, on occasion, be 
 produced by. the decomposition of 18 parts by weight of water. 
 
 From the purely chemical point of view this apparent anomaly 
 is explained by the circumstances that of the 36 parts of water 
 present on the left-hand side of equation (2) only one-half — i.e., 
 18 parts by weight — are actually decomposed into hydrogen and 
 oxygen, the other 18 parts remaining unattacked, and merely 
 attaching themselves as "water of hydration" to the 56 parts of 
 calcium oxide in equation (1) so as to produce 74 parts calcium 
 hydroxide appearing on right-hand side of equation (2). When 
 the output of gas is measured in terms of the water decomposed, 
 in no commercial apparatus, and, indeed, in no generator which can 
 be imagined fit for actual employment does that output of gas ever 
 approach the quantitative amount, but the volume of the water used, 
 if not actually disappearing, is always vastly in excess of the 
 requirements of equation (2). On the contrary, when the make of gas 
 is measured in the terms of the calcium carbide consumed, a 
 
30 MODERN METHODS OF WELDING 
 
 percentage may be reached of 80, 90, or even 99 per cent, of what 
 is theoretically possible. Inasmuch as calcium carbide is the only 
 costly ingredient in the manufacture of acetylene so long as it is not 
 wasted — so long, that is to say, as nearly the theoretical yield of 
 gas is obtained from it — an acetylene generator is satisfactory or 
 efficient in this particular ; and, except for the matter of solubility, 
 the quantity of water consumed is of no importance whatever. 
 
 The chemical action between calcium carbide and water is accom- 
 panied by a large involution of heat, which, unless due precautions 
 are taken to prevent it, raises the temperature of the substances 
 employed, and of the apparatus containing them, to a serious and 
 often inconvenient extent. This phenomenon is the most important 
 of all in connection with acetylene manufacture, for upon a proper 
 recognition of it, and upon the character of the precautions taken 
 to avoid its numerous evil effects, depend the actual value and 
 capacity for smooth working of an acetylene generator. Just as, 
 by an immutable law of chemistry, a given weight of calcium carbide 
 yields a given weight of acetylene, and by no amount of ingenuity 
 can be made to produce either more or less, so, by an immutable law 
 of physics, the decomposition of a given weight of calcium carbide 
 by water, or the decomposition of a given weight of water by calcium 
 carbide, yields a definite quantity of heat — a quantity of heat which 
 cannot be reduced or increased by any artifice whatever. 
 
 A very little experiment will show that a notable quantity of heat 
 is set free when calcium carbide is brought into contact with water, 
 and, by arranging the details of the apparatus in a suitable manner, 
 the quantity of heat manifested may be measured with considerable 
 accuracy. A lengthy description of the method performing this 
 operation is unnecessary. It is sufficient to say that the heat is 
 estimated by decomposing a known weight of carbide by means of 
 water in a small vessel surrounded on all sides by a carefully jacketed 
 receptacle full of water, provided with a sensitive thermometer. 
 The quantity of the water contained in the outer vessel being known, 
 and its temperature having been noted before the reaction com- 
 mences, an observation of the thermometer after the decomposition 
 is finished, when the mercury has reached its highest point, gives data 
 which show that the reaction between water and a known weight of 
 calcium carbide produces sufficient in amount to raise a known weight 
 of water through a known therm ometric distance ; and from these 
 figures the corresponding number of calories may easily be calculated. 
 
 It is well to remark that there is scarcely any feature in the 
 generation of acetylene from calcium carbide and water — certainly 
 
ACETYLENE 31 
 
 no important feature — which introduces into practice principles 
 not already known to chemists and engineers. Once the gas is set 
 free, it ranks simply as an inflammable, moisture-laden, somewhat 
 impure, illuminating and heat-giving gas, which has to be dried, 
 purified, stored, and led to the j)lace of combustion. It is in this 
 respect precisely analogous to coal-gas. Even the actual generation 
 is only an exothermic, or heat-producing, reaction between a solid 
 and a liquid, in which the rise in temperature and pressure must be 
 prevented as far as possible. Accordingly, there is no fundamental 
 or indispensable portion of an acetylene apparatus which lends itself 
 to the protection of the patent laws. 
 
 Treatment of Acetylene after Generation. — The calcium carbide 
 manufactured to-day, even the best obtainable, is by no means a 
 chemically pure substance. It contains a large number of impurities, 
 or foreign bodies, some of which evolve gas on treatment with water. 
 To a certain extent, this statement will always remain true in the 
 future, for in order to make absolutely pure carbide it would be neces- 
 sary for the manufacturer to obtain and employ perfectly pure lime, 
 carbon, and electrodes in the electric furnace which did not suffer 
 attack during the passage of a powerful current, or he would have to 
 devise some process simultaneously or subsequently removing from 
 his carbide those impurities which were derived from his impure raw 
 material, or from the walls of his furnace. Besides the impurities 
 thus inevitably arising from the calcium carbide decomposed, how- 
 ever, other impurities may be added to the acetylene by the action 
 of a badly designed generator, or one working on a wrong system of 
 construction. Therefore it may be said at once that the crude gas 
 coming from the generating plant is seldom fit for immediate con- 
 sumption. It must invariably be submitted to a rigorous method 
 of chemical purification. 
 
 Combining together what may be termed the carbide impurities 
 and the generator impurities, in crude acetylene the foreign bodies 
 are partly gaseous, partly liquid, partly solid. They may render the 
 gas dangerous from the point of view of possible explosion. They 
 may be harmful to health if inhaled, injurious to the fittings and 
 decorations of rooms, objectional at the blowpipe orifices by deter- 
 mining or assisting the formation of the solid growths which dis- 
 tort the flame and so reduce its power ; they may give trouble in the 
 pipes by condensing from the state of vapour in the bends and dips, 
 or by depositing, if they are already solid, in angles, etc., and so 
 causing stoppages. 
 
 It will be apparent without argument that a proper system of 
 
32 MODERN METHODS OF WELDING 
 
 purification is one that is competent to remove the carbide impurities 
 from acetylene, as far as removal is desirable or necessary. It should 
 not be necessary to extract generator impurities, because the proper 
 way to deal with them is to prevent their formation. Vapour of 
 water almost always accompanies acetylene from the generator, 
 this being due to the fact that in a generator where the carbide is in 
 excess the temperature tends to rise until part of the water is vapor- 
 ised and carried out of the decomposing chamber before it has an 
 opportunity of reacting with the excess of carbide. In large plants 
 the extraction of the moisture may take place in two stages. The 
 gas from the generator is generally passed slowly through a condenser, 
 although in smaller generators it is often quite suitable for the gas 
 to pass through the water of the generator in order to remove the 
 soluble impurities. The generator impurities present in the crudest 
 acetylene consist of hydrogen and nitrogen- — i.e., the main con- 
 stituents of air; the various gases — liquid and semisolid bodies — 
 which are produced by the polymerising and decomposing action of 
 heat upon the carbide, water, and acetylene in the apparatus ; and, 
 wherever the carbide is in excess in the generator, some lime in the 
 form of very fine dust. This lime dust, which is calcium oxide or 
 hydroxide, carried forward by the stream of gas in extremely fine 
 subdivision, is liable to be produced whenever water acts rapidly 
 upon an excess of calcium carbide. It occasionally appears in the 
 alternative form of froth in the pipes leading directly from the 
 generating chamber. This froth is hard to break up. 
 
 A purifying system must remove generator impurities, unless the 
 generator is so perfect that it does not give them off. With the 
 exception of the gases which are permanent at atmospheric pressure 
 — hydrogen, carbon monoxide, nitrogen, and oxygen — which, once 
 produced, must remain in the acetylene, extraction of these im- 
 purities is quite simple. The dust or froth of lime will be removed 
 in the washer where the acetylene bubbles through water. The 
 dust itself can be extracted by merely filtering the gas through 
 cotton-wool, felt, or the like. The least volatile liquid impurities 
 can be removed partly in the condenser, partly in the washer, partly 
 by a mechanical dry-scrubbing action of the solid purifying materials 
 in the chemical purifier. To some extent the more volatile liquid 
 bodies may be removed similarly. 
 
 Carbide Impurities. — Neglecting very minute amounts of carbon 
 monoxide and hydrogen as being insignificant from the practical 
 point of view, the carbide impurities of the gas fall into three mam 
 categories : those containing sulphur, those containing silicon, those 
 
ACETYLENE 33 
 
 containing gaseous ammonia. The phosphorus in the gas becomes 
 calcium phosphide in the calcium carbide which is attacked by 
 water, and yields phosphuretted hydrogen (or phosphine). The 
 calcium phosphide, in its turn, is produced in the electric furnace 
 by the action of the coke upon the phosphorus in phosphatic lime. 
 The sulphur in the gas comes from aluminium sulphide in the carbide. 
 Even in the absence of aluminium compounds, sulphuretted hydro- 
 gen may be found in the gas of an acetylene generator. In the gas 
 itself the ammonia exists as such, the phosphorus mainly as phos- 
 phine. The sulphur is present partly as sulphuretted hydrogen, 
 partly as organic compounds analogous, in all probability, to those 
 of phosphorus. Ammonia and sulphuretted hydrogen are both 
 soluble in water, the latter more particularly in limewater of an 
 active acetylene generator, while all other bodies referred to are 
 completely insoluble. Therefore a proper washing of crude gas 
 in the water should suffice to remove all ammonia and sulphuretted 
 hydrogen from the acetylene. 
 
 When acetylene was first introduced on commercial lines, 
 the generator manufacturers began to attack the problem of 
 purification in a perfectly empirical way, either employing some 
 purely mechanical scrubber, filled with moist or dry porous material, 
 or perhaps coke or the like, wetted with dilute acid. At first sight 
 it might appear that the methods of treating coal-gas would suit 
 acetylene, as the latter contains two of the impurities, sulphuretted 
 hydrogen and ammonia. Setting on one side, as worthy of atten- 
 tion, certain compositions offered as acetylene-purifying materials, 
 whose constitutions have not been developed, and whose action 
 has not been certified by respectable authority, there are now three 
 principal chemical reagents in regular use. These are chromic acid, 
 cuprous chloride, and bleaching powder. Chromic acid is employed 
 in the form of solution acidified with acetic acid. In order to obtain 
 the advantages attendant upon the use of a solid purifying material, 
 this is absorbed in that highly porous and inert silica known as 
 infusorial earth or " kieselguhr." This substance was first recom- 
 mended by Ullmann, and is termed commercially " heratol." As 
 sold, it contains about 136 grammes of chromic acid per kilogramme. 
 Cuprous chloride is used as a solution in strong hydrochloric 
 acid, mixed ferric chloride, similarly absorbed in " kieselg hr." 
 From the name of its proposer, this composition is called "franko- 
 line." It will be observed that both heratol and frankoline are 
 powerfully acid, whence it follows that they are capable of extract- 
 ing any ammonia that may enter the purifier. But this material 
 
34 MODERN METHODS OF WELDING 
 
 should be in an earthen vessel. Heratol changes somewhat in colour 
 as it becomes spent, its original tint, due to the chromic acid, altering 
 to a dirty green, characteristic of the reduced state of the chromium. 
 
 Frankoline has been asserted to be capable of regeneration or 
 revivification — i.e., when spent it may be rendered fit for further 
 service by being exposed to the air for a time, as is done with gas 
 oxide. 
 
 Of all these materials, heratol is the completest purifier of 
 acetylene, removing phosphorus and sulphur most rapidly and 
 thoroughly, and not appreciably diminishing in speed or efficiency 
 until its chromic acid is practically used up. On the other hand, 
 heratol does not act on pure acetylene, so that purifiers containing 
 it should be small in size, and frequently recharged. 
 
 Frankoline is very efficacious as regards phosphorus, but it does 
 not extract sulphur. The purifying materials mentioned may be 
 valued by their price, proper allowance being made for the quantity 
 of gas purified per unit weight of substance taken. The annexed 
 table shows approximately : 
 
 (1) The number of litres of gas purified by 1 kilogramme of 
 the substance, and 
 
 (2) The number of cubic feet purified per pound. 
 
 
 Litres per 
 
 Cubic Feet 
 
 
 Kilogramme. 
 
 per Pound, 
 
 Heratol 
 
 5,000 
 
 80 
 
 Frankoline 
 
 9,000 
 
 144 
 
 Puratylene 
 
 1,000 
 
 156 
 
 Opinions differ as to the maximum volume of acetylene which a 
 certain variety of purifying material will treat. If 1 pound of a 
 certain substance will purify 200 cubic feet of normal crude acetylene, 
 that weight is sufficient to treat the gas evolved from 40 pounds 
 of carbide, but it will only do so provided it is so disposed in the 
 purifier that the gas does not pass through it at too high a speed 
 that it is capable of complete exhaustion. 
 
 Purifiers charged with heratol are stated, however, to admit of a 
 more rapid flow of the gas than is the case with other materials. 
 The ordinary allowance is 1 pound of heratol for every cubic foot 
 per hour of acetylene passing, with a minimum charge of 7 pounds 
 of the material. As the quantity of the material is increased, the 
 flow of the gas per hour may be proportionately increased — e.g., 
 a purifier charged with 133 pounds of heratol should purify 144 
 cubic feet of acetylene per hour. 
 
CHAPTER VII 
 OXYGEN CYLINDERS 
 
 The rapid and extensive developments of the oxy-acetylene process 
 in recent times has led, of course, to a corresponding growth in the 
 handling and use of oxygen cylinders. If simple and indispensable 
 precautions are followed in manipulating the cylinders and utilising 
 the gas they contain, no real danger is present; but many welders 
 are entirely ignorant of the care required, and the author proposes 
 therefore to mention a few principal points in the construction and 
 handling of these cylinders. 
 
 The manufacture, transport, and utilisation of cylinders of com- 
 pressed gases were investigated by a Government Committee in 
 1895, and, although their recommendations have not been converted 
 into law, they form the basis for the manufacture of cylinders and 
 the regulation of the gas-cylinder trade. The oxygen cylinders 
 usually employed in oxy-acetylene welding contain from 100 to 200 
 cubic feet of gas under a pressure of 120 atmospheres, or 1, 800 pounds 
 per square inch. The cylinders containing 100 cubic feet are most 
 used, the approximate dimensions of this size being 7 inches 
 diameter and 49 inches over-all length, including valve. The cylinder 
 weighs ajDproximately 1 cwt. The majority of the cylinders are 
 made of seamless steel, and the method of manufacture is of interest. 
 A flat steel slab, about f inch thick, is raised to a red heat, and sub- 
 jected to three hot drawing-through processes. The cylinder is 
 next annealed, and then pickled in acid to remove the scale. After 
 this it is subjected to about six cold drawings, so as to produce the 
 right shape, length, and thickness. The bottom of the cylinder is 
 hemispherical and the open end is swaged down, after previously 
 upsetting the end of the tube so as to form the neck, which, in turn, 
 is screwed to receive the cylinder's valve. 
 
 After manufacture, the cylinders are annealed by the manu- 
 facturers, and reannealed, valved, and tested by the compressing 
 firm. The cylinders are reannealed every third or fourth year, and 
 retested every one or two years. They are stamped on the neck, 
 so that at any time in their history they can be traced. The most 
 important marks are the annealing and test-mark, giving the date 
 
36 MODERN METHODS OF WELDING 
 
 of the last testing and annealing. A useful mark on the cylinder 
 is that of its capacity or internal volume. A foot is sometimes 
 shrunk on the base of the cylinder. This enables it to be placed 
 in an upright position, and avoids sudden shocks in manipulation. 
 Sectional views through typical oxygen cylinders are shown in 
 Fig. 9. The cap is shown separately. 
 
 With regard to the testing of cylinders, ordinary tensile, elonga- 
 tion, and bending tests can only be carried out on finished cylinders 
 by destroying them. It is useful to carry out tests to destruction 
 on about 2 per cent, of the cylinders made at one time to any given 
 specification. Tests for cylinders which are to be put into use are 
 of two kinds : 
 
 (1) The Hydraulic Test. — The cylinder is subjected to an hydraulic 
 pressure, usually about double the intended working pressure. 
 If a cylinder stands this test, it should be perfectly safe for the work- 
 ing pressure. The hydraulic test gives no information as to whether 
 the cylinder is annealed or not or whether the steel is ductile. 
 
 (2) The Stretch Test. — It being necessary to examine whether 
 the cylinder has permanently stretched during the hydraulic test, 
 compressing firms have introduced an easily applied and sensitive 
 test called the stretch test. During the hydraulic test the cylinder 
 is placed in a vessel, which forms a water-jacket. A pipe and glass 
 tube communicate with this jacket. If there is any expansion of the 
 cylinder during the hydraulic test water is driven out of the jacket 
 and rises in the glass tube. Oxygen cylinders always contain a 
 certain amount of water which it is impossible to eliminate, and 
 which accumulates during successive fillings, finally occupying 
 several cubic inches. It is therefore necessary carefully to drive 
 out the water contained in a full cylinder before screwing the pres- 
 sure-reducing valve in position. In this manner the life of the 
 cylinder is increased, and the difficulties encountered when using 
 large quantities of oxygen (as, for example, in cutting) are removed 
 or reduced. 
 
 Cylinders of compressed oxygen can be manipulated without any 
 very special precautions. For use they are placed, according to 
 shape, upright or lying down. Carefully avoid letting them fall. 
 Such accidents do not affect cylinders, but may injure the welders, 
 and in many cases damage the reducing valve. The cock or valve 
 i£ the delicate part of the cylinder. The welder has only, in theory, 
 to open and close the valve at the beginning and the end of the 
 work. This operation, so simple in itself, requires certain precau- 
 tions. Often on the arrival of the cylinder the valve is hard to 
 
OXYGEN CYLINDERS 37 
 
 open. The operator should make sure of its working before placing 
 the reducing valve in position, so that any powdered oxide or other 
 dust is blown away. The oxygen on escaping into the air produces 
 a violent hissing. The valve should be opened and closed alternately 
 two or three times. The slightest escape can be detected by the 
 ear. The reducing valve should now be fixed, care being taken 
 to see that the screw on the valve is put in square to the thread of 
 the cylinder, so as to avoid damaging the thread. Screw down 
 tightly with the spanner supplied with the cylinder. Then open 
 the tap on the reducing valve where the rubber pipe fits on, and turn 
 on gently the gas at the cylinder tap, and test if there is any leakage 
 at the cylinder or valve. If there is an escape, which may be shown 
 by the pointer on the gauge rising after the valve of the cylinder has 
 been closed, try to screw the valve tighter without overdoing it, 
 because it will be difficult to reopen after. If the leak still con- 
 tinues, remove the reducing valve, and open the cylinder valve 
 briskly two or three times. 
 
 No grease, oil, soap, or any fatty matter must be used, as 
 oxygen under pressure has an oxidising action on all these articles. 
 This causes heat to be produced, which may start combustion, and 
 the conflagration may spread to the ebonite parts of the valve and 
 destroy them. The oxygen then escapes in large quantities from 
 the cylinder, thus tending to produce a brisk combustion by con- 
 tact with a lighted body. The results may be serious. In case the 
 leak is considerable and it is possible to stop it by ordinary methods, 
 the cylinder should be returned to the manufacturers, with a label 
 attached, "Valve faulty," so that the manufacturers can proceed 
 to repair it before refilling. 
 
 There are cases of freezing of the reducing valve by the solidifica- 
 tion of water -vapour. Particular warning must be given against 
 melting the ice by heating with the flame of the blowpipe. This 
 is a bad practice, and may lead to accidents. The only thing is to 
 use warm water. In order to avoid excessive expansion of the gas, 
 and the resulting increase in pressure, the cylinders should always 
 be kept away from a warm place. Avoid placing them in the sun 
 or near fires. After emptying, the cylinders should be immediately 
 returned to the company to be refilled. They generally belong to 
 the manufacturers, but one can purchase one's own. Oxygen 
 cylinders are carried on the railways at class 2 rates by goods 
 train, and empties returned at reduced rates. All sent by rail 
 must be fitted with covers, as specified by the Railway Clearing 
 House. 
 
38 MODERN METHODS OF WELDING 
 
 Although much literature on oxy-acetylene welding and cutting 
 has appeared in recent years, there has not been any of a general 
 character in which concise and comprehensive regulations have been 
 given to blowpipe operators. 
 
 One frequently comes across cases of trained and efficient welders 
 who are incurring daily risks at their work simply because many 
 important precautions and regulations are unknown to them. It 
 is safe to say far more welding accidents occur through ignorance 
 than through wilful neglect of ordinary safeguards. Only a very 
 small proportion of operators have had the advantage of training at a 
 welding school. No opportunity should be lost of placing regulations 
 bearing on their own security before the large number of men and 
 women operators now employed throughout the country. I append, 
 therefore, the necessary regulations, and I impress on all operators 
 to carry them out for the safeguarding of themselves and others. 
 
 Regulations, Precautions, and Safeguards. 
 
 (1) In placing contracts for oxygen supplies, a guarantee from 
 the supplier should be obtained, to the effect that oxygen will only 
 be delivered in cylinders which have been- made, annealed, tested, 
 and filled strictly in accordance with the recommendations of the 
 British Government Departmental Committee of 1896. 
 
 The British railways and all the responsible road and water 
 carriers require these conditions to be complied with. 
 
 (2) A guarantee should be also obtained to the effect that the 
 oxygen supplied will be not less than 98-5 per cent, quality. 
 
 (3) See that all cylinders supplied with oxygen are painted black 
 and fitted with right-hand valves ; never attempt to alter the colour 
 of the cylinder or the screwed thread of the valve connections. 
 
 (4) See that cylinders are not exposed to excessive heat. 
 
 (5) Oxygen cylinders should not be exposed to temperature 
 exceeding 1,000° F. 
 
 (6) Take care not to lay hot welded or cut material on or along- 
 side oxygen cylinders ; and great care must be taken never to allow 
 the blowpipe flame or the heat from the same to impinge on the 
 oxygen cylinder. 
 
 (7) Carefully avoid the use of oil or grease, or lubricant in any 
 form, upon the cylinder valves or fittings, and keep same dry and 
 free from grit. 
 
 (8) Never use keys of long leverage to close cylinder valve; 
 they give under power, which is injurious to the valves, and fre- 
 
OXYGEN CYLINDERS 
 
 39 
 
 quently results in broken spindles. If the valve leaks when closed 
 with ordinary lever key, it is often due to grit. To remove this, 
 open the valve slowly, and then close it sharply. 
 
 (9) After disconnecting a cylinder before it is empty, it 
 is desirable to test for leakage 
 by pouring some water into the 
 valve socket. If no bubbles ap- 
 pear in the water, it proves that 
 the valve is gas-tight. The valve 
 gland can be tested in the same 
 way at any time that the valve 
 is open, and a regulator at- 
 tached, by pouring water into 
 the recessed part of the gland- 
 nut round the spindle. The 
 gland-nut must be tightened up 
 if necessary to prevent leakage. 
 
 (10) See that all socket and 
 nipple ends are in good order and 
 free from grit before fixing regu- 
 lators or other fittings by screwing 
 up the fly-nut " hand tight." 
 
 (11) Never open the cylinder 
 valve suddenly when the regulator 
 is fixed. The valve should be 
 opened by tapping the key gently 
 with the hand. 
 
 (12) Store cylinders under 
 cover in an outhouse, or in a 
 portion of the premises most 
 remote from any source of fire 
 risk. Avoid as much as possible 
 exposing cylinders to conditions 
 which promote oxidisation, and 
 return empty cylinders to the 
 oxygen factory as quickly as 
 
 possible. Cylinders should be returned with a label bearing 
 the customer's name, so that they may be identified and credited 
 to him in the oxygen factory. 
 
 Fig. 9.- 
 
 -SeCTION OF AN OXYGEN; 
 
 Cylinder. 
 
CHAPTER VIII 
 ACETYLENE GENERATORS 
 
 General Construction. — Acetylene generators may be divided 
 into several classes. For the purpose of description three types 
 will be taken — namely, water-to-carbide, carbide-to-water, and the 
 dipping or contact generators. We will describe each one by itself, 
 and state the advantages and disadvantages as they arise. 
 
 Most people, when taking up oxy-acetylene welding, decide to 
 purchase a small portable generator. This is a mistake. Such a 
 generator is only intended for occasional use. It frequently 
 happens, therefore, that the purchaser finds after a few weeks that 
 the generator is not large enough, and he has to purchase a larger 
 one. It is very much better to have a large one at first. The first 
 cost may be higher, but the advantages are many, and the saving in 
 gas would pay for the extra cost in a few months. 
 
 Generators should be well designed by competent men who have 
 had years of experience, and everything should be carefully thought 
 out and well constructed, somewhat on the following lines : 
 
 (1) Generating casing should be made of strong steel sheets, and, 
 after manufacture, galvanised. 
 
 (2) The tubes and fittings all galvanised. 
 
 (3) Two or more generating chambers. 
 
 (4) Generator capable of recharging while working. 
 
 (5) Quite automatic. 
 
 (6) Capable of taking any over generation. 
 
 (7) Washer to absorb the ammonia and sulphurous acid. 
 
 (8) Purifier with chemical composition for the removal of phos- 
 phorus and other impurities. 
 
 (9) Large outlet and inlet tubes, so as not to throttle the gas. 
 
 (10) No copper should be used on any part of the generator. 
 The production of acetylene by the action of water on calcium 
 
 of carbide is, chemically, one of the simplest of reactions. But in 
 practice it is not so simple. The two chief difficulties in the pro- 
 duction of acetylene are heating and excess production. Water 
 consists of hydrogen and oxygen, the dissociation of which takes 
 
 40 
 
ACETYLENE GENERATORS 41 
 
 place with the absorption of heat. On the other band, the oxygen 
 liberated combines with calcium carbide and produces the action 
 more heat than is absorbed by the above reaction. The heat liber- 
 ated is 226 calories or 900 B.T.U. per pound of carbide — that is to 
 say, 1 pound of carbide would raise 1 gallon of water from 0° to 50° C. 
 No device or arrangement can alter this amount of heat liberated; 
 but the temperature of the mass altered will not go beyond the 
 temperature of boiling water. This result is caused in two ways: 
 (1) The water vaporises and acts with the carbide, thus supple- 
 menting the heat of decomposition; (2) under the influence of heat 
 the lime gives up the water, reaction continues, and if there is no 
 external cooling the temperature rises. The generation of acetylene 
 at a high temperature is detrimental, and causes polymerisation. 
 This is one reason why manufacturers often put the generating 
 chambers inside the tanks, so as to have the water all round them, 
 and also to have the outlet pipes from the generating chambers 
 turned down from the top, so that the gas is passed through the 
 water cooled and washed. 
 
 Polymerisation, as stated previously, takes place owing to too 
 rapid generation, which results in excessive heat to a temperature 
 of 130° F. and causes the lime or spent carbide to turn yellow in the 
 carbide trays, being in the form of a tarry substance. 
 
 Carbide gives up sulphide on the action of heat; and the water 
 decomposes into hydrogen sulphide and organic sulphur compounds, 
 which are very detrimental to acetylene. The unpleasant odours of 
 sulphur dioxide are given off on the heating of the gas. 
 
 No generator produces the proper proportion of gas consumed. 
 The production, therefore, is either in excess or deficient. Produc- 
 tion being in advance, the sudden stoppage of consumption cannot 
 possibly correspond to the above arrest of the reaction. 
 
 It is important to note that any particular generator-heating and 
 after-generation are in direct relation to the delivery. It is there- 
 fore impossible to formulate rules on this point without taking into 
 account the delivery. The heating should never exceed a tempera- 
 ture of 130° C. The generator bell should be large enough to take 
 all the after-generation in the case of stoppage. A well-designed 
 and constructed generator needs the minimum of attention. It is 
 entirely automatic, only gives gas as required, is strong, and will 
 last for a number of years. 
 
 The characteristic point in every generator for welding is its 
 flexibility. It should adapt itself to fluctuating employment — 
 that is, should rise to the maximum and minimum yield without 
 
42 MODERN METHODS OF WELDING 
 
 delay, without heating, without jerks. This refers to automatic 
 generators only. 
 
 Non-automatic generators are usually of large dimensions, in 
 which the gas is made in large quantities in advance and stored in a 
 gasometer of from 50 to 500 cubic feet. This method is the most 
 practical, most sure, and most economical by a long way, but the 
 initial cost is high. 
 
 Fig. 10. — 10-Cwt. Carbide-to-Water Generator. 
 
 Fig. 10 is a photograph of a large generating plant, comprising 
 generator, condenser, washer, gasometer, and purifier. It is 
 worked on the usual principle. 
 
 Figs. 11 and 12 are carbide-to- water type generators, medium- 
 pressure type, which makes it possible for the storage of the 
 generated gas, with its accompanying advantage of absolute 
 volumetric control. The gasometer obviates the variation of gas 
 
ACETYLENE GENERATORS 
 
 43 
 
 pressure that is inherent in the pressure generator. No regulating 
 or reducing device is necessary, as the acetylene is generated at the 
 pressure required for use. The possibility of loss of gas through 
 leakage in the line and connections is practically eliminated with 
 the low-pressure system. The gas bell provides storage for gas and 
 effectively guards against the loss due to after-generation inherent 
 in other systems. 
 
 Control of the feed 
 mechanism is accomplished 
 through the rise and fall of 
 the gas bell, and with abso- 
 lute gas-pressure. The move- 
 ment of the gas bell gives 
 a dual motor control — first, 
 through a brake; second, 
 through a position jaw 
 clutch. Special provision is 
 made for shutting off the 
 motor feed when the gas 
 bell is in its lowest position. 
 
 Operation of the genera- 
 tor is effected by a small 
 but efficient weight-driven 
 motor, which automatically 
 starts and stops to supply 
 the amount of gas being 
 used. The motor weights 
 always lower approximately 
 the same distance for each 
 pound of carbide used, and 
 constitutes a reliable indica- 
 tion of the amount of carbide 
 remaining in the machine at 
 any time. By the use of 
 
 positive forced feed, it is impossible for more than the proper 
 quantity of carbide to be fed in the water. In securing cool, 
 and hence efficient, generation, it is necessary to have not less 
 than 1 gallon of water capacity per pound of carbide charge. It 
 is impossible for the temperature of the gas to rise above the 
 boiling-point of water, the acetylene"! bubbling through the water, 
 free from some of the impurities. By the time the gas is ready 
 to leave the surface outlet its temperature does not exceed that 
 
 Fig. 11. — 50 Pounds Carbide Capacity, 
 
 using 1J Pounds Carbide. 
 
 Height, 62 inches; diameter, 24 inches; 
 
 weight, 450 pounds. 
 
44 
 
 MODERN METHODS OF WELDING 
 
 of the air by more than a few degrees. These conditions permit 
 a yield of pure gas. 
 
 A complete and efficient system of interlocking safety devices 
 prevents mistakes in operation due to carelessness or forgetfulness 
 when charging the generator. The hydraulic back-pressure valve 
 
 is of unique design. With three 
 distinct water seals, it prevents 
 the possibility of any oxygen 
 entering the generator. The 
 generator is equipped with an 
 agitator that churns the resi- 
 duum thoroughly, allowing it 
 to flow freely and quickly when 
 the residuum is opened for re- 
 charging. 
 
 These generators produce 
 acetylene at a pressure of less 
 than 15 pounds. After the 
 hopper has been filled with 
 carbide (l|-inch mesh) it is fed 
 to water by means of a re- 
 volving disc; surrounding this 
 disc are little plows, or scrapers, 
 that scrape off a certain quan- 
 tity of the carbide as the disc 
 revolves. The disc is controlled 
 by a governor which is actuated 
 by the gasometer. When the 
 gasometer reaches a predeter- 
 mined r eight, the motor is 
 stopped by means of a brake. 
 When the gas bell recedes, the 
 motor is automatically started 
 by releasing the brake. The 
 gas passes from the generator 
 chamber into the gasometer. 
 From there it passes through 
 the filter, where physical impurities and suspended matter are 
 removed, and then through the hydraulic valve, which is a water seal 
 for preventing the reverse flow of gas or air. The gas then passes 
 into the service line. The filter consists of a cylindrical shell within 
 which felt is placed. Perforated plates are placed at the top and 
 
 Fig. 12. — 100 Pounds Carbide Capacity, 
 using 1J Pounds Carbide. 
 
 Height, 76 inches; diameter, 30 inches; 
 weight, 550 pounds. Known as Davis- 
 Bournonville Pressure Generators. 
 
ACETYLENE GENERATORS 45 
 
 bottom. The gas passes up through the perforations and felt, which 
 collects all dirt, lime, or portions of sludge that would otherwise 
 enter the service line with the gas. 
 
 The bell tank is divided by the horizontal partition, the lower 
 portion of the tank forming the seal chamber. The stand pipe 
 projects downward into this chamber, being sealed with water it 
 contains. The function of the seal chamber is to prevent an ex- 
 cessive pressure from accumulating within the apparatus, and pre- 
 cludes the possibility of a backward or reverse pressure entering the 
 generator. Should the pressure of the gas from any cause be in- 
 creased above the normal pressure, the seal will immediately break. 
 The water will overflow from the seal, filling lip, and release the 
 pressure. 
 
 If there should be generated, because of defective or broken 
 apparatus or other reasons, a larger volume of gas than can readily 
 be held by the gas bell, the bell would be forced upward. In order 
 to prevent such an occurrence, there is a stand pipe arranged with a 
 series of blow-off holes. This stand pipe is connected to the vent 
 pipe. Should the stand pipe rise above the level of the water, thus 
 producing a free passage for the gas down the stand pipe through 
 the safety vent pipe to the outside air, as soon as the gas bell was 
 relieved of the excess gas it would lower gradually and the blow-off 
 holes would be again under the water and the flow of gas shut off, 
 thus permitting the generator to resume its normal working condition. 
 The vent valve, the generator filling valve, and the residuum gate 
 are so arranged that they can only be operated in sequence. The 
 opening of the residuum gate therefore allows the gas within the 
 generator to escape to the outside air before any valve or opening 
 from the generator can be made into the room. 
 
CHAPTER IX 
 OXYGEN REGULATORS 
 
 Oxygen-reducing valves are manufactured by various firms, and 
 are of various forms. Some are very reliable and, with care, will 
 last a good number of years. The reducing valves are fixed to the 
 oxygen cylinder by means of a union on the valve. The screws on 
 the union are standardised, as likewise the screws on the top of the 
 cylinder. In the valve of the cylinder there is a faced surface 
 
 Fig. 13. — Oxygen Regulators: 1 Two Gauges, 1 One Gauge, 1 No Gauge. 
 
 (concave) ; on the regulator tip is a convex-faced and ground surface. 
 In fixing the regulator tight in the cylinder, the cylinder spanner 
 should be used, and care taken to see that it is tight ; before turning 
 the oxygen on, open the tap of the blowpipe supply, then gently 
 open the valve on the cylinder and see if all is sound. Turn off the 
 blowpipe tap. The regulator must not leak at the cylinder; if it 
 does, turn off the oxygen again and tighten up the union. Try the 
 gas on again. If still leaking, the regulator should be entirely 
 removed, and the oxygen turned on two or three times to blow out 
 any grit or dust that may have accumulated inside the valve. 
 
 46 
 
OXYGEN REGULATORS 
 
 47 
 
 Then fix the regulator again, and test. This time you will probably 
 be sound. 
 
 It is most important to note that there is great danger in getting 
 oil or grease into the union of the reducing valve, because ignition 
 may take place and cause an explosion, doing much damage. 
 
 There are three types of regulators ; one is shown in Fig. 13. No. 1 
 has no gauge and, naturally, the cost is less owing to its absence. It 
 is not absolutely necessary to have this gauge. The regulator works 
 
 ^ #>^ 
 
 Fig. 14. 
 
 -Double Cylinder Connector, in which One can be Used while 
 the Other Empty One can be Changed. 
 
 quite as well without, but one cannot tell what amount of gas there 
 is in the cylinder; it has got to be used till empty. No. 2 is the same 
 as No. 1, but has a gauge fitted, which registers the gas in the cylinder. 
 No. 3 is for high pressure and is used for cutting purposes. One 
 gauge is for the pressure which is regulated by the tee screw. The 
 other indicates the amount of gas in the cylinder. All oxygen 
 cylinders are painted black, and have a right-hand thread for fitting 
 into the cylinders. These regulators are suitable for every class of 
 work for which oxygen and other compressed gases are used. They 
 automatically deliver gas from the cylinders at any pressure to which 
 they are set. This is very important in welding and for the correct 
 
48 MODERN METHODS OF WELDING 
 
 mixture of the gases. They are substantial in construction, are 
 fitted with a gas expansion device which obviates ignition risks at the 
 valve seat, and are specially recommended for use for all kinds of 
 blowpipe work in connection with oxygen cylinders. The adjust- 
 able screwed socket on the side of the regulator No. 1 is graduated 
 in pounds per square inch. The regulator can be set by this to 
 any desired constant pressure. 
 
 No. 2 regulator has a high-pressure gauge to register the cylinder 
 pressure ; but these pressure gauges, permanently attached to regula- 
 tors, are a fruitful source of trouble. They soon become inaccurate 
 (particularly the small type so frequently employed), and, being 
 delicate in construction, are liable to injury in workshop handling. 
 The connector illustrated in Fig. 14 is an excellent substitute for the 
 pressure gauge. The regulator is in communication with the cylin- 
 ders A and B, one of which can be cut off when not in use. Thus, 
 if the valve of the cylinder A and the pipe valve a are open, whilst 
 the valve of cylinder B and the valve b are closed, oxygen flows 
 from cylinder A through the regulator till it empties. The valve 
 of cylinder A is then closed and that of cylinder B opened. Oxygen 
 will then flow to cylinder B, whilst the empty cylinder A can be 
 removed and replaced by a full one. It will readily be seen that 
 a continuous supply of oxygen can be maintained by the employ- 
 ment of this connector. For prolonged use, regulators with these 
 connectors will be found more convenient and more reliable than 
 those fitted with pressure gauges. 
 
 The oxygen should be opened as slowly as possible on to the 
 regulator, and the regulating screw should be fully open. This 
 avoids the heating by quick compression. This is important for the 
 safety of the welder, and preserves the regulator in good working 
 order, because sudden pressure on the diaphragm of the regulator 
 produces derangement and often puts it out of order through the 
 diaphragm splitting. Regulation of the pressure should be secured 
 by the regulating screw until the pressure is that stamped on the 
 blowpipe to be used, and the outlet valve should be full open. 
 
CHAPTER X 
 
 REGULATIONS 
 
 Regulations, Precautions, and Safeguards for Oxy- Acetylene Weld- 
 ing and Cutting. — There lias been much literature on oxy-acetylene 
 welding and cutting in the past, but little has been said as to the 
 precautions necessary. All students and others interested in the oxy- 
 acetylene welding and cutting should study these regulations very 
 closely. One frequently comes across cases of trained and efficient 
 welders who are incurring daily risks at their work simply because 
 many important precautions and regulations are unknown to them. 
 It is safe to say that far more welding accidents occur through ignor- 
 ance than wilful neglect of ordinary safeguards. 
 
 The general regulations which it is necessary to observe in 
 connection with oxygen cylinders have been given above. I now 
 add similar regulations with regard to carbide and to acetylene 
 generators. 
 
 Carbide. 
 
 (1) Carbide must be stored in iron or steel vessels, hermetically 
 sealed. 
 
 (2) The vessels should be kept in a dry and well-ventilated place. 
 
 (3) No artificial light capable of igniting inflammable vapour 
 should be employed near these vessels nor in any room where carbide 
 is stored. 
 
 (4) Carbide can only be stored without a licence in a quantity 
 not exceeding 28 pounds. 
 
 (5) If it is desired to store larger quantities a licence must be 
 obtained from the local authorities. 
 
 (6) Carbide should be of a quality to yield not less than 4-8 cubic 
 feet of acetylene per pound. 
 
 Fig. 15 shows an airtight carbide chamber; it holds 2 cwt. 
 
 Acetylene Generators. 
 
 (1) See that the generator is of ample capacity for the continuous 
 production, without heating, of the maximum quantity of acetylene 
 required, and that it complies with all official recommendations. 
 
 49 4 
 
50 
 
 MODERN METHODS OF WELDING 
 
 (2) See that, whether the system employed be automatic or non- 
 automatic, the holder is of sufficient capacity to obviate any loss of 
 gas due to production when the supply to the blowpipe is cut off. 
 
 (3) See that the design precludes any appreciable admission of 
 air to the apparatus in the charging with carbide. 
 
 (4) See that the limit of pressure in any part of the apparatus 
 does not exceed 250 inches of water. 
 
 (5) See that the size of the pipes conveying the gas is propor- 
 tioned to the maximum rate of generation. 
 
 Fig. 15. — Airtight Carbide Chamber. 
 
 (6) See that it is impossible to seal hermetically the generating 
 apparatus. 
 
 (7) See that no copper fittings are employed in connection with 
 the acetylene apparatus. 
 
 (8) See that all back-pressure valves are in working order as per 
 regulations. 
 
 (9) Charge the apparatus with carbide, if possible, only by day, 
 and do not use small-grained carbide. 
 
 (10) Keep the apparatus clean and in good order, and carefully 
 remove all sludge from the generator before recharging with carbide. 
 
 (11) Do not use naked lights in the vicinity of the acetylene 
 apparatus. 
 
 (12) In frosty weather never use a stove in the vicinity of the 
 acetylene generator. To prevent freezing of water, it is best to 
 employ a steam or hot-water coil. 
 
REGULATIONS 51 
 
 (13) Employ suitable purifying material and recharge after 
 1 10 cubic feet per pound of purifying material have passed through 
 it; test occasionally the acetylene issuing from the blowpipe with 
 a piece of nitrate of silver paper. If this is at all discoloured, it 
 indicates that the purifier requires to be recharged. 
 
 (14) See that, in all cases where an acetylene generator is em- 
 ployed, an hydraulic back-pressure valve is employed between it 
 and every blowpipe in use, and that it is properly filled with water. 
 The small drain tap should be opened after filling, any excess of 
 water being drained off. The hydraulic valve must be tested every 
 day in this way before being used. 
 
 (15) If, through a back-fire or other cause, water is discharged 
 from the vent pipe, always refill the chamber, and see that the small 
 drip tap is afterwards opened to draw off any excess of water. 
 
 (16) The hydraulic back-pressure valve should be dismantled 
 at regular intervals and cleaned out, to make sure that the vent pipe 
 and passages are clear. 
 
CHAPTER XI 
 BLOWPIPES 
 
 Blowpipes intended for the autogenous welding of metals, which 
 employ oxygen and acetylene as the gases, are manufactured instru- 
 ments, with the same accuracy as a watch ; and they must be used 
 as such with care. They are light and easy to handle, and are made 
 with great skill so as to allow the correct proportions of acet3dene 
 and oxygen, in specified measured volumes, at a fixed velocity, 
 
 Section of Fouche biowpipe 
 
 Fig. 16. — Fotjche Blowpipe, Section and Elevation. 
 
 according to the size of the blowpipe. Their length and weight 
 vary, as does their size, ranging from a consumption of 1-75 of acety- 
 lene to 100 cubic feet per hour. 
 
 The consumption of acetylene should be about 1 to 1-3 of oxygen. 
 The necessary conditions are much more difficult to realise than they 
 appear to be, especially with blowpipes in which the acetylene is 
 admitted at the pressure of generation, which is about 8 to 12 inches 
 water pressure. A great difficulty is in obtaining the requisite 
 stability. A large number of details merit attention, such as ease 
 of manipulation and ease of taking to pieces and reassembling. 
 
 52 
 
BLOWPIPES 
 
 53 
 
 The velocity of propagation is about 330 feet per second in the case 
 of oxygen and acetylene. In order to avoid the striking back of the 
 flame it is necessary that the velocity of the mixture at the exit 
 
 » fl 
 
 pq 
 
 should be of the same value, so that it prevents the return of the 
 flame to the interior. The oxygen being under pressure, it is easy 
 to keep this gas constant; but acetylene is not under pressure, and 
 the blowpipes have to be designed to get the amount required to 
 complete the correct mixture for combustion. 
 
54 
 
 MODERN METHODS OF WELDING 
 
 Low-pressure blowpipes are designed on the injector principle, 
 and have separate internal jets for the oxygen fixed inside the blow- 
 pipe. Such a jet has a very small hole bored in it, the size being 
 determined by the size of the blowpipe for which it is to be used. 
 
 Fig. 19. — Multiple Tips Universal Blowpipe, Small Size. 
 
 On the outside of this inner oxygen jet is space for the acetylene, 
 which is attached to a tube (usually) which runs along the pipe, where 
 the rubber tubing is fixed. The oxygen under pressure rushes out 
 through the small internal jet, in the place where the acetylene is; 
 
 Fig. 20. — Multiple Tips Universal Blowpipe, Large Size. 
 
 and the velocity of the oxygen draws the acetylene with it to the 
 mixing chamber and out of the outer nozzle, in the proper propor- 
 tion required for a good steady flame. Most injector blowpipes 
 are designed on this principle. The intimate mixtures of the gases 
 
BLOWPIPES 55 
 
 should be perfectly accomplished before they escape from the blow- 
 pipe. This is difficult to attain, because it is necessary to avoid too 
 much loss of pressure, which would demand an increase in the pres- 
 sure of oxygen in order to regain the required A^elocity at the exit. 
 This would be detrimental to the weld. 
 
 A blowpipe, which in appearance is such a common article, 
 requires such precision in construction that it can only be under- 
 taken by specialists in the subject. It is essential, indeed, to leave 
 to specialists and experts not only the design and construction but 
 even the repair of blowpipes. Economy and good results in weld- 
 ing depend largely on this. Delivery of oxygen being fixed by the 
 size of the injector orifice, and the power of the blowpipe being 
 invariable in these limits, therefore, in practice, variation of pres- 
 sure clearly means bad welding. The makers stamp on each size 
 
 Fig. 21. — Endazzle Blowpipe, Single Tip Pattern. 
 
 of blowpipe the correct pressure at which it will work and give the 
 best results. This pressure should not be increased. It is quite a 
 common practice among welders, when their blowpipes are not 
 working well, to increase the oxygen pressure with the idea of getting 
 a better flame. This is a very bad practice indeed, and the weld is 
 usually spoilt. Too much importance cannot be laid on this vital 
 point. Manufacturers who are expert in their line would not stamp 
 a working pressure on their blowpipe if it can be used for higher 
 pressure. 
 
 According to the different thicknesses of the metal to be welded, 
 various sizes of blowpipes will be wanted. The pressures and 
 volumes of gases required varying with the size of the welds, it is 
 necessary, therefore, to have blowpipes designed to suit. They 
 range, as has been said, from T5 to 100 cubic feet per hour of acety- 
 lene gas. There is a great variety of blowpipes at present on the 
 market — -some good, some medium, and some bad. All operators 
 should make themselves fully conversant with the various designs 
 and the manufacture of the same. One thing that must be remem- 
 bered is that the orifice of the nozzle of any blowpipe is proportionate 
 
56 
 
 MODERN METHODS OF WELDING 
 
 to the delivery of the injector when using the pressure of oxygen 
 stipulated by the makers. It is essential that it should not be 
 reduced or enlarged. If it were, the gases would not be correctly 
 mixed for the proper combustion for a stable flame. Almost 
 the first blowpipes for low-pressure welding were made by 
 Fouche, a Frenchman. These were well constructed and very 
 
 Fig. 22. — Endazzle Blowpipe, Multiple Tip Pattern. 
 
 reliable in working. In fact, they are still more reliable than 
 many more modern ones. Their only fault was that they were 
 heavy. Fig. 16 shows one in section and one in elevation. 
 
 The "Universal" blowpipes are made by the British Oxygen Com- 
 pany. They are very largely used, and are good blowpipes. They 
 are standardised, and new parts for renewals can easily be got . The 
 
BLOWPIPES 
 
 57 
 
 universal blowpipes may have either a fixed or interchangeable 
 head. The two kinds are practically the same in appearance and 
 construction. In the interchangeable set there are loose heads of 
 various sizes, with one handle only. The heads vary in power, 
 and are numbered to correspond with the proper pressure required. 
 
 **H 
 
 Fig. 23. — Osborne Blowpipes, Four Different Types. 
 
 The small siza is supplied with seven heads ranging from 2 to 8, 
 and the larger size with four heads, representing 8, 10, 12, 15. 
 These blowpipes are compact and useful. 
 
 The blowpipe shown on p. 55, known as the "Endazzle," is ex- 
 tremely light, and has a unique attachment: a pressed-steel hinged 
 cover over the rubber tube connectors. These open right out to 
 allow the rubber tubes to be fixed on the ends of the blowpipe, and 
 
58 
 
 MODERN METHODS OF WELDING 
 
 then afterwards close up, which prevents the hands from being 
 burnt should ignition take place at the handle of the blowpijae. 
 This often occurs if the tubing is not a good fit. 
 
 The type below (Figs. 24 and 25) — injector heads with tips com- 
 plete — is made in several sizes for operating on different sections of 
 metals. For greater convenience these blowpipes are made in two 
 models. Each model is supplied complete with directions, and each 
 injector head is of proper proportions to produce the correct mixtures 
 of gases and a flame of perfect stability and correct dimensions 
 according to the work for which it is intended. Model A has a 
 range of injector heads — sizes to 4 — that is, five different blow- 
 pipes (heads only). Model B has a range of injector heads — 1 to 
 9 — 'that is, nine different blowpipes (heads only). This covers a good 
 range and will weld anything from T \ inch to 1 inch thick. 
 
 Fig. 24. — Small Style C Welding Blowpipe, No. 453. 
 
 These blowpipes, known as the Davis-Bournonville type (Figs. 
 24 and 25), are provided in two standard sizes — large and small — 
 which can be fitted with either square or angle heads (45, 75, 
 or 90 degrees), and straight or angle hose connections; but the 
 standard model is shown here. This blowpipe is very desir- 
 able for light and medium sheet metal welding and light repair 
 work, where a light, compact, nicely balanced tool is appre- 
 ciated. Weight, 18 ounces; length over all, 14 inches. Fitted 
 with five tips — Nos. 1, 2, 3, 4, 5, style 99 — using oxygen pressures 
 of 2, 4, 6, 8, and 10 pounds respectively. It is used to advantage 
 on metal ^V to T 5 g inch thick. 
 
 A standard blowpipe for heavy welding, and for general 
 shop work requiring a strong blowpipe. Weight, 2 pounds; 
 length over all, 20 inches. Fitted with five tips — Nos. 6, 7, 
 8, 9, 10, style 100 — using oxygen pressures of 12, 14, 16, 18, and 
 
BLOWPIPES 
 
 59 
 
 20 pounds respectively. This blowpipe can be used on metal 
 from J inch thick upward. 
 
 Consumption of Blowpipes. — Blowpipes of high, medium, and 
 low pressures are constructed so as to give flames of all intensities 
 requisite for the practice of autogenous welding. The power is 
 reckoned according to the hourly consumption of acetylene. 
 Some consume 1-7 to 100 cubic feet of acetylene per hour, or 20 
 cubic feet, which corresponds to the hourly delivery of acetylene 
 The blowpipe with the lowest consumption uses about 1 -5 cubic feet 
 per hour of acetylene. This welds sheet iron or steel up to T V mcn 
 thick; larger blowpipes consume 80 to 100 cubic feet of acetylene 
 per hour, and these weld 1-inch thick material. 
 
 In dealing with the consumption of acetylene, it is as well that 
 we should deal with the oxygen at the same time. Theoretically 
 
 Fig. 25. — Large Style C Welding Blowpipe, No. 146. 
 
 we know that to get a correct mixture and a neutral flame with a 
 small white cone the gases must be mixed in equal proportions — 
 that is, 1 volume of oxygen, 1 volume of acetylene, when measured 
 under normal temperature. Blowpipes for the high and medium 
 pressures obtain this result by using the two gases under equal 
 pressures, direct from the two cylinders — oxygen and acetylene. 
 It is nearly approached in blowpipes in the medium-pressure 
 system; but it cannot be reached in the low-pressure system. 
 
 It is general, in ordinary works practice, to use 1 of acetylene 
 to 1-3 of oxygen; it is only in well-designed blowpipes, working under 
 normal conditions, well-regulated flame, without apparent excess of 
 either oxygen or acetylene, and an expert welder, that these results 
 are obtained. But low-pressure blowpipes give trouble if they are 
 not well taken care of to prevent their getting knocked about. 
 Also there are the difficulties with the regular supply of acetylene 
 
60 MODERN METHODS OF WELDING 
 
 and the constant pressure. The oxygen being under pressure, 
 it is difficult to mix the acetylene in absolutely accurate quantities 
 in order to give it sufficient velocity. 
 
 There is evidently an energetic mixing of the two gases, but the 
 contact is not molecule to molecule, and the stream lines of oxygen 
 or acetylene can escape at the nozzle without being mixed. One can 
 test this in different ways — for example, by contracting the exit 
 tube of the mixing chamber, or by increasing the pressure of the 
 oxygen. In both cases the proportion of oxygen to acetylene is 
 raised considerably. From this it is apparent that blowpipes for 
 low pressure use least oxygen when the admission pressure of oxygen 
 is least, and this arrangement for obtaining a mixture of the two 
 gases is the best. In some blowpipes the oxygen pressure to be used 
 corresponds with the arrangement of the mixing chamber. Change 
 of section and abrupt bending produce a loss of pressure. One 
 must find an equilibrium between the two factors, which are opposite. 
 If the pressure of oxygen is not raised too high, the arrangement of 
 mixing is excellent, and the result will be perfect. From prac- 
 tice it is well known that, as the welding proceeds, the blowpipe 
 becomes heated, and the gases, especially acetylene, expand. This 
 causes a decrease in acetylene gas, and makes the flame at once 
 oxidising. 
 
 From tests which have been made upon the consumption of the 
 gases by the Congress on Autogenous Welding, getting fifty blow- 
 pipes from the different manufacturers (the average delivery of 
 acetylene was fixed at 350 litres per hour, and the work to be executed 
 lasted from thirty to forty minutes: the welders were experts), 
 the best proportion of oxygen to acetylene was 1-12, the average 
 1-3, and the worst 1-9. A test was also made with the same blow- 
 pipe handled by two welders, one using oxygen at 28 pounds pressure 
 and the other at 13 pounds. The proportion in the first case was 
 1*83, in the second 1-25, showing clearly the influence of excess 
 pressure of oxygen on the consumption of the gas. These tests 
 prove that, according to the type of blowpipe and the conditions of 
 use, the consumption of oxygen for a constant delivery of acetylene 
 can vary greatly and may double in volume. Not only is the oxygen 
 consumed in excess of the theoretical amount a pure loss; its pres- 
 ence in the flame oxidises the metal, lowers the strength of the weld, 
 and renders it brittle and porous. These considerations are im- 
 portant from the point of view of economy and good work, and those 
 interested should carefully study them. 
 
 In the choosing of blowpipes many things are to be taken into 
 
BLOWPIPES 61 
 
 account; if it is to be used for continuous work on one thickness of 
 metal, then the proper sized blowpipe should be chosen with a fixed 
 delivery. Then, again, one must satisfy oneself that one is buying 
 the best article, not so much as regards appearance or shape, as 
 with a view to lowest consumption for the particular size of work; 
 and a guarantee should be got from the makers for a stipulated 
 hourly consumption. If the class of welding is changeable from 
 thick to lighter materials, then a combined independent set with all 
 interchangeable heads would be most suitable. This applies to 
 small equipments in small shops. On the other hand, in large shops 
 where a good number of operators are empkyyed it is more economi- 
 cal to have fixed ones ; they are not so delicate as the interchange- 
 able, which, by the constant changing, suffer more wear. 
 
 The actual weight is often important in practical use. Some- 
 times welders say that the best blowpipes are those that are light 
 in the hand; but unless they have attended instructional classes 
 they have not the slightest knowledge of the consumption, or other 
 details which must be settled before purchase. If the work is 
 continuous, then a light blowpipe should be adopted, providing, of 
 course, that the working is right, with the correct mixtures of gases 
 and the consumption up to standard. If the work is heavy, such 
 as some repairs which have to be done quickly, then a heavy type 
 would be preferable. 
 
 The questions of working, the consumption of the gases, and 
 the maintenance in the workshop, have been badly neglected in the 
 past. As competition is getting keener daily, however, manu- 
 facturers are now interesting themselves in the details. One comes 
 across many blowpipes which are well constructed and regulated, 
 but have that tiresome striking back of the flame into the interior 
 when the nozzle gets heated. This is a serious defect, because the 
 welder generally increases the pressure of oxygen. 
 
 A guarantee should be got from the suppliers that this back-firing 
 will not take place. The matter of consumption is a vital point as 
 regards economy and cost. In large shops, where there are one 
 hundred or more blowpipes in use at once, the saving in oxygen, 
 with the very best designed blowpipes, giving a consumption of 
 1*3 of oxygen to 1 of acetylene, may amount to hundreds of pounds 
 per year. 
 
 The author has made tests in this direction, one of which may 
 be described here. Six blowpipes were used, two each of different 
 manufacture, which we call A and Al, B and Bl, C and CI. These 
 tests were made on a 1-inch plate, butt-welded, with 12-gauge thick 
 
62 MODERN METHODS OF WELDING 
 
 charcoal iron wire as the welding-rod. These tests lasted twenty- 
 five minutes, and the following were the results : 
 
 A and Al gave 1-3 of oxygen to 1 of acetylene. 
 B „ Bl „ 14 „ „ „ 
 
 C ,, CI ,, 1-75 f , ,, ,, 
 
 These are clear instances of the varying makes of blowpipes, 
 and it at once demonstrates how important it is to have the very 
 best designed blowpipes that are made. As a further illustration : 
 suppose a bad blowpipe, using 1 -75 of oxygen to 1 volume of acety- 
 lene; say that the consumption of the works is 4,000 cubic feet 
 per month— that is, ten cylinders of 100 cubic feet each per week 
 (many firms use this quantity per hour). Under these conditions 
 it would be — ■ 
 
 ^-^ = 2,971 cubic feet. 
 
 The loss on a badly designed blowpipe is, therefore, 1,029 cubic 
 feet, which, at Id. per foot, is £4 5s. 9d. per month, and £51 9s. per 
 year. This is only 10 cylinders per week — for larger users, the saving 
 is greater. Apart from the loss, this excess of oxygen is highly 
 detrimental to the welds, which is much more serious even than the 
 loss of the gas. 
 
 Maintenance of Blowpipes. 
 
 Users of blowpipes must bear in mind always that they are 
 articles of precision. They are made on delicate lines, to be deli- 
 cately used, and not to hammer the weld as the author has seen 
 some operators do. If carefully protected, they will last for years, 
 just the same as when new. The taking to pieces must not be done 
 with cumbersome tools, and in cleaning the nozzles, copper wire 
 should be used. If the orifice is in any way enlarged, the slightest 
 alteration in section produces derangement. The blowpipe section 
 of the nozzle corresponds to a determined flow of oxygen, but the 
 orifice for the flowing of the oxygen (the injector) remains unchanged, 
 and any increase of the nozzle opening brings about a decrease of the 
 velocity at the exit, which provokes a return of the flame into the 
 interior of the blowpipe. When welding, the oxides, or particles of 
 metal, produce the following results : 
 
 The delivery of the oxygen being variable and escaping under 
 greater pressure than the acetylene, causes the flame to become 
 oxidising in effect. The orifice being smaller for the passages of the 
 
BLOWPIPES 
 
 63 
 
 two gases, it is the stronger (the oxygen) that gets through in pre- 
 ference to the acetylene. 
 
 One must never allow oil or grease on the blowpipes or tubes, 
 as in oxygen the oil may catch fire, and usually burns the rubber 
 tube. This often happens to new blowpipes, in which the oil has 
 got inside during manufacture. Do not take a blowpipe to pieces 
 unless you are versed in its component parts, especially the inner 
 
 UB 
 
 Fig. 26. — Showing the Correct Neutral Flame, Middle One Correct. 
 
 jet. This can rarely be adjusted again hi the same place. Before 
 sending out they are adjusted to a gauge, and one-thousandth of 
 an inch out in this injector sets it all wrong. The best and quickest 
 method is to return it to the makers for repairs. 
 
 If a blowpipe is obstructed by dust or other particles (generally 
 lime dust carried over from the generator), these should be got away 
 by fixing the rubber tube on the nozzle end of the blowpipe, leaving 
 
64 
 
 MODERN METHODS OF WELDING 
 
 the taps open and turning on the oxygen temporarily. This should 
 blow it quite clean. 
 
 All operators should take great pride in blowpipes trusted to 
 their care, and should, as I have said before, treat them as instru- 
 ments of precision, keeping them ranged in good order and always 
 polished. 
 
 The following are a few hints which will be found useful : 
 
 (1) See that the blowpipe is in good order, and no passages 
 obstructed; also that the rubber tubes are correct and securely 
 fixed, and the regulator on the oxygen cylinder in proper working 
 order. 
 
 (2) See that you have ample acetylene and oxygen for the 
 work in hand before commencing; it is injurious to the weld to stop 
 in the middle. 
 
 (3) Turn the oxygen and acetylene taps full on, and light the 
 blowpipe. The flame will then probably have an excess of acetylene, 
 which should be reduced by gradually turning the acetylene tap 
 on the blowpipe (or the outlet of the hydraulic valve, if there is no 
 tap on the blowpipe), until the flame of the blowpipe has a clearly 
 defined cone at the orifice. Fig. 26 shows what is required. The 
 first one has an excess of acetylene; the second is correct; and 
 the third has too much oxygen. 
 
 Blowpipes required for welding by the oxy-acetylene system 
 must be chosen with care, must come up to the standard rules, and 
 must not use more than 1-3 volumes of oxygen to 1 of acetylene. 
 Also they must be easy of regulation, easy to handle, and able to keep 
 up a regular and stable flame over long periods of working. 
 
 The following is an approximate table, giving the consumption 
 of each size of blowpipe, for oxygen and acetylene; the size of the 
 blowpipe to use, with the thickness of the plate being welded, and 
 the length of feet that should be welded per hour. These tables 
 are very useful and should have close attention. 
 
 Size of blowpipe 
 
 Approximate thickness of plate- 
 joint 
 
 Approximate con-' 1 . 
 
 sumption of[/ oxygen .. 
 gases per hour! ^acetylene .. 
 in cubic feet . . J 
 
 Feet welded per hour 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 10 
 
 12 
 
 — 
 
 — 
 
 — 
 
 i" 
 
 8 
 
 — 
 
 4 
 
 3* 
 
 8 
 
 X" 
 
 3" 
 4 
 
 1-7 
 1-2 
 
 3 
 2 
 
 6-5 
 4-3 
 
 9 
 6-3 
 
 16 
 11 
 
 23 
 16 
 
 34 
 
 24 
 
 48 
 34 
 
 75 
 48 
 
 40 
 
 30 
 
 20 
 
 15 
 
 12 
 
 9 
 
 7 
 
 H 
 
 2| 
 
 15 
 
 1" 
 
 100 
 
 70 
 
 H 
 
CHAPTER XII 
 FLEXIBLE TUBING 
 
 No welding installation is complete without the means to conduct 
 the gas from the oxygen cylinders and the hydraulic valves. The 
 gas is conveyed by rubber tubing fixed on the proper connectors, 
 attached to, firstly, the blowpipe; secondly, the hydraulic safety 
 valve; and thirdly, the regulator on the oxygen cylinder. This 
 rubber tubing is very important in any installation, and unless great 
 care is taken by the welders, and the best quality of rubber purchased, 
 it is an expensive maintenance charge. A cheap quality of rubber 
 tubing is useless. Some of that on the market contains much more 
 loading material than rubber. Such is dear at any price, although 
 in appearance there is not much to choose. The best tubing con- 
 sists of a good, soft, pliable inside rubber liner, of good stout thickness 
 and not less than f inch inside clear diameter, reinforced by at 
 least three-ply heavy woven canvas. It must suit the connectors 
 on the blowpipes, regulators, and hydraulic safety valves. The fit 
 should be just sufficient to secure a gas-tight joint at the connectors, 
 but not too tight, so that it can be removed without undue strain 
 or stretching of the tubing. There are 60 lineal feet in each coil 
 of rubber tubing, and these usually cut into four pieces, making the 
 requirements for two operators for two 15 feet each, one for the 
 acetylene and one for the oxygen. 
 
 This rubber tubing is badly abused in the workshop. It is often 
 blown open by the welders suddenly turning the oxygen on full at 
 high pressure when the blowpipe tap is closed. Also at times they 
 burn the tubing whilst welding, unknown to themselves. Itmayget 
 accidentally thrown across the hot article which is being welded, 
 and usually this is not found out till a hole has been burnt in it. 
 Again, the tubing frequently gets cut by articles dropped on to it. 
 On making an examination on the tubing being used, it will often 
 be found that one sample in five is leaking, and the oxygen blowing 
 away in the atmosphere. This is a costly item and should be 
 watched very closely. Often, too, the tubing ignites at the connec- 
 tor of the blowpipe, because the rubber liner is curled up inside 
 
 65 5 
 
66 MODERN METHODS OF WELDING 
 
 and does not make a gas-tight joint; consequently, the oxygen 
 escapes. The welder, not aware of this, continues welding till a 
 spark flashes from the weld and ignites the rubber tubing at the 
 connector of the blowpipe. It becomes incandescent immediately, 
 probably burning the operator's hand, unless he is quick enough 
 to drop the lighted blowpipe. It is necessary to have tubing of the 
 correct size to fit the connectors, so as to avoid the bad practice of 
 tying with wire. 
 
 Neither the connector nor the rubber should have any grease 
 or oil on it. This will set up instant combustion if any oxygen 
 catches it. If the end of the tubing is hard to get on, it should be 
 dipped in water. This will ease the fixing on of the connectors. 
 All connections, unfortunately, are not alike, which makes the fitting 
 of tubing awkward where the connectors vary. It would be a great 
 boon if all manufacturers of these connections were to standardise 
 the sizes. It would cheapen production, and save much time lost 
 in trying to get one size tubing on another size connector. For 
 cutting blowpipes, the rubber tubing must be very much stronger, at 
 least five-ply, owing to the greater pressure required for cutting iron 
 and steel. 
 
 In many cases, when cutting very thick plate at high pressures, 
 armoured tubing must be used. 
 
 The connectors on the blowpipes, regulators, hydraulic safety 
 valves, should all be painted with shellac : this assists in keeping the 
 tubing on without using wire, and makes a tight joint. 
 
 The present system of connectors for the rubber tubing is 
 not very satisfactory, and new ones ought to be brought into service. 
 These new ones consist of a union joint with coupling, so that one 
 half of the union can be fitted on the gauge on the cylinder, and the 
 other half fitted on the rubber tubing at the gauge end. These 
 union couplings should also be attached to the hydraulic safety valve, 
 and at the other end of the tubing to that which is attached to the 
 gauge on the cylinder, and the same union couplings on the blowpipe. 
 The great advantage is that half the coupling is fixed permanently 
 on each end of the tubing, thereby saving much time and preserv- 
 ing tubing. 
 
CHAPTER XIII 
 SAFETY VALVES 
 
 It is a well-known fact that oxygen and acetylene are very highly 
 explosive, and it is indispensable that all precautions should be 
 taken to prevent the formation of these combustible gases. Their 
 use is becoming more and more general in every part of the country, 
 and we have therefore to put forward continually that precautions 
 must be taken to prevent their formation, especially as their inflam- 
 mation is very easily produced. With acetylene in use at a lower 
 pressure than oxygen, the oxygen can return in the acetylene tubes 
 and piping and so combine this gas in the generator. This occurs 
 when there is a partial obstruction of the blowpipe, caused often by 
 the welder allowing the blowpipe suddenly to touch the molten metal. 
 The oxygen, being under pressure, and as the outlet is blocked, 
 flows up the acetylene tube into the safety valve or hydraulic seal. 
 
 Therefore it is absolutely essential to place in the acetylene 
 piping, between the generator and the rubber acetylene tubes on the 
 blowpipes, an arrangement capable of arresting immediately any 
 return of the oxygen. The object of a safety valve is to direct into 
 the open air any oxygen which returns in the direction of the acety- 
 lene. It is not really meant to stop back-fire, but to prevent forma- 
 tion of a mixture of high explosive gases by the return of the flame. 
 
 The efficacy of such an apparatus must be absolute. So far a 
 simply designed water seal has proved the most effective. It is one 
 that cannot go wrong if the water level is kept right. In the hydrau- 
 lic safety valve, two tubes emerge from a layer of water, one for the 
 entry of the gas and the other open to the exterior, placed at different 
 levels, constituting an absolute barrier against all return of the 
 oxygen in the acetylene piping. Other arrangements, not based 
 on this principle, should be rejected. 
 
 The illustration (Fig. 27) shows a good standard type made by 
 the British Oxygen Company. The acetylene pipe from the gas- 
 holder or main supply is connected with tap A, and the acetylene 
 tube leading to the blowpipe is connected with tap B. C is a loosely 
 fitting lid, covering the cup in which the water is poured to charge 
 
 67 
 
68 
 
 MODERN METHODS OF WELDING 
 
 the seal pot D up to the level of the tap E. Taps A and B must be 
 closed whilst the seal pot D is being charged with water. When 
 water shows at tap E, immediately stop filling, and, allowing time 
 for the surplus water to drain off, close the tap E. The lid C must 
 
 Fig. 27. — Safety Valve, Elevation and Sectional. 
 
 then be replaced, and the taps A and B may be opened. The valve 
 is then in working order. 
 
 The filling pipe F is made long enough to hold a column of water 
 greater than the pressure of the acetylene generator. There should 
 not be less than 8 inches of water. When at work the taps A and B 
 must be open, and the supply of acetylene regulated by the tap on 
 the blowpipe. Should the blowpipe nozzle at any time become 
 
SAFETY VALVES 
 
 69 
 
 choked whilst the oxygen supply remains unchecked, the gas would 
 be forced by its superior pressure along the acetylene tube. The 
 back pressure thus caused, acting on the surface of the water in the 
 seal pot D, would seal the acetylene pipe and force the water up the 
 pipe F, displacing the liquid 0. The hydraulic seal to the atmo- 
 
 Fig. 28. — Standard Sectional Type Safety Valve. 
 
 sphere would thus be destroyed and both gases would escape until 
 taps A and B were closed. Thus oxygen can never penetrate the 
 acetylene supply beyond the hydraulic valve, provided the valve 
 is kept properly filled. 
 
 There is really very little scope for modification in the design 
 of these hydraulic valves. Two conditions, however, are essential 
 
70 MODERN METHODS OF WELDING 
 
 — viz., that the pipe conveying acetylene from the generator dis- 
 charges into the water seal pot at a lower level than the opening to 
 the vent ; and that the vent pipe has a direct vertical discharge from 
 the seal pot into the atmosphere. 
 
 All hydraulic back-pressure valves must be fixed on the acetylene 
 supply pipe in a vertical position, as near to the blowpipe as con- 
 venience will permit. A good position is on a wall, with the bottom 
 of the valve about 4 feet from the ground. The acetylene inlet pipe, 
 coupled to the tap A, should extend vertically several feet above the 
 tap. In cases where two or more blowpipes are worked from the 
 same acetylene supply a separate hydraulic back-pressure valve 
 should be employed for each. 
 
 One must be very careful in the choice of hydraulic back-pressure 
 valves. A serious explosion, resulting in grave injuries to more than 
 one workman and the total destruction of an acetylene generator, 
 occurred in a munition works in Yorkshire in May, 1917. The holder 
 formed part of a large plant which had been in successful operation 
 for several years. Suspicion not unnaturally attached itself to the 
 hydraulic valve. 
 
 No part of an oxy-acetylene outfit is more important or requires 
 more careful attention than the hydraulic back-pressure valve. 
 The explosion in question was caused by a badly designed hydraulic 
 back-pressure valve of German make, in which there was a U-tube 
 fitted, one end of which was fixed in the seal pot, and the difference 
 between the U-pipe and the supply pipe was not sufficient to make a 
 satisfactory seal. It is obvious that such a valve as this is useless 
 as the means of preventing oxygen, at its superior pressure, from 
 flowing back to the acetylene generator. The essential method of 
 working is for the acetylene to bubble through a small height of 
 water, which is nevertheless sufficient for covering the tube leading 
 to the exterior, this being between the surface of the water and the 
 level of the escaping acetylene. The valves must not be too large 
 a gas capacity. The diameter of the body should just be suffi- 
 cient to retain the level of the water constant, and the height enough 
 to avoid drops of water reaching the outlet of the acetylene. The 
 pipe which comes from the main supply into the seal pot should be of 
 suitable diameter for maximum delivery to the largest blowpipe, 
 so as to avoid all loss of pressure. 
 
 Often a pipe of small diameter, when a large blowpipe is used, 
 causes eddies in the flow of the gas, through the delivery not being 
 sufficient. The pipe which leads from the generator should go 
 through the seal pot to within \ inch of the bottom. The bottom 
 
SAFETY VALVES 71 
 
 end of this should be shaped in the form of a cone, and the cone should 
 have small holes drilled in it to allow the gas to spread more when 
 passing through the water. The height of the water in the seal pot 
 should be about 3 inches clear of the top holes in the cone. This 
 will be the position of the test tap. The acetylene outlet tap will 
 be about 8 inches above this. The atmosphere pipe should be placed 
 half-way between the test tap and the holes of the inlet acetylene pipe. 
 The height of this pipe depends on the pressure of the acetylene, 
 since the water rises in this tube as the pressure of the gas is in- 
 creased. The height should be related to the level of the water in 
 the valve, and should be more than the greatest possible pressure 
 that would be used in the generators. 
 
 The illustration (Fig. 28) shows a valve that is infallible in working 
 and is simple in construction. It can easily be made by any in- 
 telligent man. It consists of one piece of solid drawn tube, 3 inches 
 inside diameter, with discs welded top and bottom. The bottom 
 disc should have a quarter gas socket welded on, to take a tap to 
 drain the water out when this is required for cleaning purposes. 
 Gas sockets may also be welded at each of the tap holes for screwing 
 them in. The acetylene inlet tube and the filling tube may both be 
 welded in the disc top before welding the top on ; but one must be 
 careful to get the tubes fixed at the proper depth before welding 
 them in. 
 
 The outside tube terminates in a funnel, which is used for filling 
 the valve with water. This is covered by a lid. Well-designed and 
 well-constructed hydraulic valves work well, are safe, and do not 
 get out of order. It is only necessary to verify the level of the water 
 daily, or every time it is left standing, and all operators should see, 
 and make a practice of, trying the test tap not less than twice a day. 
 
CHAPTER XIV 
 
 PURIFIERS 
 
 Purifiers generally consist of cylindrical vessels, usually made of 
 sheet steel with an airtight lid or cover. They usually contain a 
 series of trays holding the purifying materials. The calcium carbide, 
 as now manufactured, is by no means a chemically pure substance. 
 It includes a large number of foreign bodies. In crude acetylene, 
 these are partly gaseous, partly liquid, partly solid. They may 
 render the gas dangerous from the point of view of possible explo- 
 sions. They, or the products derived from them on combustion, 
 may be harmful to the health if inhaled. They are objectionable 
 at the burner orifices, by determining, or assisting in, the defects of 
 the metals of the weld. 
 
 A proper system of purification is one that is competent^ to 
 remove the carbide impurities from the acetylene, as far as that 
 removal is desirable or necessary. The generator impurities, as 
 stated above, are oxygen, nitrogen, and lime in the form of fine dust. 
 This lime may be extracted when the gas is passing from the genera- 
 ting chambers along the outlet pipe and down again through the bent 
 pipe which dips in the water of the tank. As the gas bubbles through 
 the water, part of the lime dust is removed. What escapes extrac- 
 tion may be removed by passing the gas through cotton- wool or felt, 
 which is usually placed over the purifying material, in the top of the 
 purifier. 
 
 The least volatile liquid impurities will be removed partly in the 
 condenser (if one is fixed), partly in the washer (the tank), and 
 partly by mechanical dry-scrubbing action of the solid purifying 
 material in the chemical purifier. Sufficient removal of these genera- 
 tor impurities need throw no appreciable extra labour upon the con- 
 sumer of acetylene, for one can readily select a type of generator 
 in which the production is reduced to a minimum, using a cotton- 
 wool or coke filter for the gas. A water washer, which is very useful 
 in the plant, if only employed as a non-return valve between the 
 generator and the main piping and the indispensable chemical puri- 
 
 72 
 
PURIFIERS 73 
 
 fiers, will take out of the acetylene all the remaining generator im- 
 purities which need to, and can, be extracted. 
 
 In designing a washer for the extraction of the ammonia and 
 sulphuretted hydrogen, it is necessary to see that the gas is brought 
 into most intimate contact with the liquid, while no more pressure 
 than can be avoided is lost. One volume of water only absorbs 
 about 3 volumes of sulphuretted hydrogen at atmospheric tempera- 
 ture, but takes up some 600 volumes of ammonia ; and, as ammonia 
 always accompanies the sulphuretted hydrogen, the latter may be 
 said to be absorbed in the washer by a solution of ammonia, a liquid 
 in which sulphuretted hydrogen is much more soluble. Since the 
 water only dissolves about an equal volume of acetylene, the liquid 
 in the washer will continue to extract ammonia and sulphuretted 
 hydrogen long after it is saturated with the hydrocarbon. To avoid 
 waste of acetylene by dissolution in the clean water of the washer, 
 the plan is sometimes adopted of introducing water into the genera- 
 tor through the washer so that, practically, the carbide is always 
 attacked by a liquid saturated with acetylene. For compactness 
 and simplicity of parts, the water of the holder seal is often used as 
 a washing liquid. But unless the liquid of the seal is constantly 
 renewed it will become offensive, and will act corrosively on the 
 metal of the tank and bell. 
 
 The reason why the carbide impurities must be removed from 
 acetylene is this: There are three compounds of phosphorus, all 
 termed phosphuretted hydrogen or phosphine — a gas PH 2 , a liquid 
 P 2 H 4 , and a solid P 4 H 2 . The liquid is spontaneously inflammable 
 in the presence of air — -that is to say, it catches fire of itself, without 
 the assistance of a spark or flame, immediately it comes in contact 
 with the atmospheric oxygen. Being very volatile, it is easily 
 carried away as vapour by any permanent gas. In commercial 
 carbide it has been found that the highest amount of phosphine in 
 the acetylene is 2-3 per cent., and this gas is capable of self -inflamma- 
 tion. Bullier states that acetylene must contain 80 per cent, of 
 phosphine to render it spontaneously inflammable. 
 
 Ammonia is objectionable in acetylene because it corrodes the 
 brass fittings and pipes, and because it is partly converted into 
 nitrous and nitric acids as it passes through the flame. 
 
 Sulphur is objectionable in acetylene because it is converted 
 into sulphurous and sulphuric anhyhrides, and their respective acids, 
 as it passes through the flame. 
 
 Phosphorus is objectionable because, in similar circumstances, 
 it produces phosphoric anhydride and phosphoric acid. Each of 
 
74 
 
 MODERN METHODS OF WELDING 
 
 these acids is harmful to the human system, sulphuric and phos- 
 phoric anhydrides (S0 2 and P 4 O 10 ) acting as a specific irritant to 
 the lungs of persons predisposed to affections of the bronchial 
 organs. 
 
 Phosphorus, however, has a further harmful action. Sulphuric 
 anhydride is an invisible gas, but phosphorous anhydride is a solid 
 body, and is produced as an extremely fine, light, white, voluminous 
 dust, which causes a more or less opaque haze. Phosphoric anhy- 
 dride is also partly deposited in the solid state at the burner orifice, 
 
 Fig. 29. — Purifier, showing Section and Elevation with Purifying Material. 
 
 and, always assisting in the deposition of carbon from any poly- 
 merised hydrocarbon in the acetylene, thus helps to block up 
 and distort the orifices of the blowpipes. 
 
 Purifiers are usually made from sheet iron or steel and galvanised, 
 and often in good plants contain a porcelain vessel for holding the 
 purifying materials, as acid from some of these act on the mild 
 steel shell, corroding it. Fig. 29 shows a purifier made from sheet 
 steel, in which all the joints are welded, and the connecting pipes 
 also welded in ; the purifying material is put into trays, the bottom 
 of which is perforated with small holes. The lowermost tray is set 
 about 3 inches from the bottom, the other trays on the top of this, 
 
PURIFIERS 
 
 75 
 
 leaving a space of about | inch between each tray. In the bottom 
 of each tray is a layer of thin felt, to prevent the purifying material 
 from passing through the holes and, secondly, to act as drier for 
 the gas. On the top of all the trays is a layer of felt or cotton-wool, 
 put in to extract the lime dust which has not been extracted by 
 the water filter, and which came over with the gases. This dust 
 is thereby prevented from getting into the blowpipe or the weld. 
 
 The purifying material should be lightly placed in the trays 
 so as to allow the gases to go freely through the material without 
 choking. It is usual for the inlet pipe to be at the bottom, and the 
 
 Fig. 30. — Atox Purifier. 
 
 outlet pipe at the top, and a water tap must be placed at the bottom 
 to allow the water from condensation to be run off from time to 
 time. It is important that this be tried frequently, as the water 
 may be sufficient to rise up, and through the purifying material, 
 thereby nullifying its properties. The purifier may be placed any- 
 where between the generator and the main piping, but it is usually 
 near the generator. 
 
 In the systematic purification of acetylene the practical question 
 arises, How is the attendant to tell when the purifiers approach 
 
76 MODERN METHODS OF WELDING 
 
 exhaustion and need recharging? Heil has stated that the purity of 
 the gas may be judged by its atmospheric flame given by a Bunsen 
 burner. Pure acetylene gives a perfectly transparent, moderately 
 dark blue flame, which has an inner cone of pale yellowish-green 
 colour, whilst the impure gas yields a longer flame of an opaque 
 orange-red tint with a bluish-red inner cone. It must be noted, 
 however, that particles of lime dust in the gas may cause the atmo- 
 spheric flame to be reddish or yellowish (through the action of calcium 
 or sodium), quite apart from the ordinary impurities. 
 
 The simple method of ascertaining, definitely, whether a purifier 
 is sufficiently active consists in the use of test papers prepared to the 
 prescription of G. Keppler. These papers are cut to a convenient 
 size, are put up in book form, and may be torn one at a time. In 
 order to test whether the gas is sufficiently purified, one of the papers 
 is moistened with hydrochloric acid of 10 per cent, strength, and the 
 gas issuing from the blowpipe, or pet eock, is allowed to impinge 
 on the moistened part. The original black or grey colour of paper 
 is changed to white if the gas contains a notable amount of impurity, 
 but remains unchanged if the gas is adequately purified. 
 
 The Keppler test papers turn white when the gas contains either 
 ammonia phosphine, siliciuretted hydrogen, sulphuretted hydrogen, 
 or organic sulphur compounds, but for carbon disulphide . that 
 change is slow. Thus the paper serves as a test for all impurities 
 likely to occur in acetylene. The paper is a specially prepared black 
 porous kind which has been dipped in a solution of mercuric chloride 
 (corrosive sublimate) and dried. These papers can be obtained, 
 put up in case with a bottle of acid for moistening them as required, 
 from E. Merck, 16 Jewry Street, London, E.C. 3, or from the usual 
 retail dealers in chemicals. 
 
 The sensitiveness of the test is quick. If a distinct white mark 
 appears on the moistened paper when it is exposed for five minutes 
 to a jet of acetylene, the latter is inadequately purified. If the gas 
 has passed through a purifier this test indicates that the material 
 is not efficient, that the purifier needs recharging. 
 
 The British Acetylene Association has issued the following set of 
 regulations as to purifying materials and purifiers for acetylene : 
 
 (1) The purifying material shall remove phosphorus and sulphur 
 compounds to a commercial degree — e.g., not to a greater degree 
 than will allow easy detection of escaping through its odour. 
 
 (2) The purifying material shall not yield any products capable 
 of corroding the gas mains or fittings. 
 
 (3) The purifying material shall, if possible, be efficient as a 
 
PURIFIERS 77 
 
 drying agent, but the Association does not consider this absolutely 
 necessary. 
 
 (4) The purifying material shall not, under working conditions, 
 be capable of forming explosive compounds or mixture. It is under- 
 stood, naturally, that this condition does not apply to the un- 
 avoidable mixture of the acetylene and air formed when charging 
 the purifier. 
 
 (5) The apparatus containing the purifying material shall be 
 a simple construction and capable of being recharged by an in- 
 experienced person without trouble. It should be so designed as to 
 bring the gas into proper contact with the material. 
 
 (6) The containers and purifiers should be made of such materials 
 as are dangerously affected by the respective materials used. 
 
 (7) No purifier should be sold without a card of instructions 
 suitable for hanging up in some prominent place. Such instructions 
 should be of the most detailed nature, and should not presuppose 
 any expert knowledge whatever on the part of the operator. 
 
CHAPTER XV 
 SELECTION AND INSTALLATION 
 
 It is not possible to give a direct answer to the question as to which 
 is the best type of acetylene generator. There are no generators 
 made by responsible firms which are not safe. Some are easier to 
 charge and clean than others. Some require more frequent atten- 
 tion. Some have moving parts less likely to fail, or none at all to 
 go wrong. There are contact apparatus on the market which appear 
 to give little trouble. There is very little to choose, from the chemi- 
 cal and physical view, between the generators now on the market. 
 A selection may rather be made on mechanical grounds. 
 
 The generator must be well able to produce gas as rapidly as 
 ever it will be required during the longest time the blowpipe may 
 be used. It must be strong and able to bear careless handling and 
 frequent rough manipulation of its parts. It must be built of sound 
 material, and galvanised after manufacture, so that it will not rust 
 in a few years. Each and every part must be accessible, and its 
 exterior visible. Its pipes for the gas must be large bore. The 
 number of cocks, valves, and moving parts must be reduced to a 
 minimum. It must be easy to clean, the waste lime must be readily 
 removed. It must be so fitted with vent pipes that the pressure 
 can never rise above that at which it is supposed to work. Appara- 
 tus that claims to be automatic should be perfectly automatic, 
 the water or the carbide feed being locked automatically before the 
 carbide store, the decomposing chamber, or the sludge cock can be 
 opened. 
 
 The generating chamber must always be in communication with 
 the atmosphere through a water seal vent pipe, the seal of which, 
 if necessary, the gas can blow at any time. All apparatus should be 
 fitted with rising holders, and the larger the better. The best place 
 for a generator is in the open air, or a simple open shed, if well venti- 
 lated. The diameter of the mains and service-pipes for an acetylene 
 installation must be such that the main or pipe will convey the 
 maximum quantity of gas likely to be required to feed properly all 
 the blowpipes which are connected to it, without an excessive actua- 
 
 78 
 
SELECTION AND INSTALLATION 79 
 
 ting pressure being called upon to drive the gas through the main 
 or pipe. 
 
 The practical question in gas distribution is, What quantity of 
 gas will a given actuating pressure cause to flow along a pipe of given 
 length and given diameter ? The solution of this question allows 
 of the diameter of the pipes being arranged so far that they carry 
 a required quantity of gas a given distance under the actuating 
 pressure that is most convenient or appropriate. In order to avoid, 
 as far as possible, expenditure and labour in repeating calculations, 
 tables have been drawn up from Morel's formulae, which will 
 serve to give the requisite information as to the proper sizes of pipes 
 to be used in the cases likely to be met with in ordinary practice. 
 
 Piping used for the distribution of acetylene must be sound in 
 itself, and the joints perfectly tight. Ordinary gas barrel is not good 
 enough. Joints for acetylene, like those for steam or high-pressure 
 water, must be made tight by using well-threaded fittings, so as to 
 secure metallic contact between pipe and socket. Acetylene service 
 should, wherever possible, be laid with a fall, which may be very 
 slight, towards a small closed vessel adjoining the gas-holder or 
 purifier, in order that water deposited from the gas through condensa- 
 tion of aqueous vapour may run out of the pipe into that apparatus. 
 Where it is impossible to secure an interrupted fall in that direction, 
 there should be inserted in the service pipe at the lowest point of each 
 dip it makes, a short length of pipe turned downwards and termina- 
 ting in a plug or sound tap, to remove the condensed water. 
 
 When all the fittings have been connected, the whole system of 
 pipes must be tested by putting it under a gas (or air) pressure of 
 9 to 12 inches of water, and observing on an attached pressure-gauge 
 whether any fall in pressure occurs within fifteen minutes after the 
 main inlet tap has been shut. The pressure required for this can be 
 obtained by weighing the holder. If the gauge shows a fall of pres- 
 sure of I inch or more in these circumstances, the pipes must be 
 examined until the leak is located, but it must never be searched 
 for with a light. Fittings for acetylene must have perfectly sound 
 joints and taps — common gas fittings will not do; the joints, taps, 
 ball sockets, etc., are not ground accurately enough to prevent 
 leakage. Fittings are now being specially made for acetylene, 
 which is a step in the right direction. 
 
 The conditions which a generator should fulfil before it can 
 be considered safe are a3 follows : 
 
 (1) The temperature in any part of the generator when run at 
 the maximum rate for which it is designed, for a prolonged period, 
 
80 MODERN METHODS OF WELDING 
 
 should not exceed 130° C. This may be ascertained by placing short 
 lengths of wire, drawn from fusible metal, in those parts of the 
 apparatus in which heat is likely to be generated. 
 
 (2) The generator should have an efficiency of not less than 90 
 per cent., which, with carbide yielding 5 cubic feet per pound, would 
 imply a yield of 4-5 cubic feet of gas for each pound of carbide used. 
 
 (3) The size of the pipes carrying the gas should be proportional 
 to the maximum rate of generation, so that undue back-pressure 
 from throttling may not occur. 
 
 (4) The carbide should be completely decomposed in the appara- 
 tus, so that the lime sludge discharged from the generator shall be 
 incapable of generating more gas. 
 
 (5) The pressure at any part of the apparatus, on the side of the 
 holder, should not exceed that of 250 inches of water, and on the 
 service side of same, or where no gas-holder is provided, should not 
 exceed 200 inches of water. 
 
 (6) The apparatus should give no tarry or other heavy condensa- 
 tion products from the decomposition of the carbide. 
 
 (7) In the use of a generator, regard should be had to the danger 
 of a stoppage of the passage of the gas, and the resulting increase 
 of pressure which may arise from the freezing of water. Where 
 freezing may be anticipated, steps should be taken to prevent it. 
 
 (8) The apparatus should be so constructed that no lime sludge 
 can gain access to any pipes intended for the passage or circulation 
 of water. 
 
 (9) The air space in a generator before charging should be as 
 small as possible. 
 
 (10) The use of copper should be avoided in such parts of the 
 apparatus as are liable to come in contact with acetylene. 
 
 (11) Notice to be fixed on the generator house door — " No naked 
 lights or smoking allowed." 
 
 (12) No repairs to, or alterations in, any part of a generator, 
 purifier, or other vessel which has contained acetylene shall be 
 commenced, nor, except for recharging, shall any such part or vessel 
 be cleaned out, until it has been completely filled with water, so as to 
 expel any acetylene or mixtures of air and acetylene which may 
 remain in the vessel and may cause a risk of explosion. 
 
 Having described various forms of the items which go to form 
 a well-designed acetylene installation, it may be useful to recapitu- 
 late briefly, with the object of showing the order in which they should 
 be placed. From the generator the gas passes into a condenser to 
 cool it and remove any tarry products. Next it enters a washing 
 
SELECTION AND INSTALLATION 81 
 
 apparatus filled with water to extract water-soluble impurities. 
 If the generator is of the carbide-to-water pattern, the condenser 
 may be omitted, and the washer is only required to retain any lime 
 froth and to act as water seal or non-return valve. If the generator 
 does not wash the gas, the washer must be large enough to act effi- 
 ciently as such, and between it and the condenser should be put a 
 mechanical filter to extract the dust. From the washer the acety- 
 lene travels to the holder. From the holder it passes through one 
 or two purifiers, and then travels to the drier and the filter. If the 
 holder does not throw a constant pressure, or if the purifier and the 
 drier cause irregularities, a governor or regulator must be added 
 to the drier. The acetylene is then ready to enter the service. 
 When the gas generally leaves the generator house, a full-way stop- 
 cock must be put in the main. 
 
 Generator Residues. 
 
 According to the type of generator employed, the waste product 
 removed therefrom varies from a moist powder to a thin cream or milk 
 of lime. Any waste product which is quite liquid in its consistency 
 must be completely decomposed and free from particles of calcium 
 carbide of sensible magnitude. In the case of more solid residues, 
 the less fluid they are the greater is the improbability (or the less 
 is the evidence) that the carbide has been wholly spent with the 
 apparatus. Imperfect decomposition of the carbide inside the 
 generator not only means an obvious loss of economy, but its pres- 
 ence among the residues makes careful handling of those essential 
 to avoid accidents, owing to a subsequent liberation of acetylene 
 in some unsuitable and, perhaps, closed situation. A residue which 
 is not conspicuously saturated with water must be taken out of the 
 generator-house into the open air and flooded with water, being left 
 in some uncovered receptacle for a sufficient time to ensure all the 
 acetylene being drawn off. A residue which is liquid enough to 
 flow should be run directly from the draw-off cock of the generator 
 through a closed pipe to the outside, where, if it does not discharge 
 into an open conduit, the waste pipe must be trapped, and a ventila- 
 ting trap provided so that no gas can blow back into the generator- 
 house. 
 
 As the acetylene is now brought through in the mains, it will 
 be distributed to the operators through an lrydraulic safety valve 
 and rubber tubing to the blowpipes. It is usual to fix the main 
 piping overhead, from which pipes are suspended, at specified dis- 
 tances, upon which pipes are attached an hydraulic safety valve. 
 
82 MODERN METHODS OF WELDING 
 
 It is necessary for each welder to have one hydraulic safety valve, 
 or water seal, for each blowpipe. There should be welding tables 
 or benches fixed running under the main piping. These tables 
 or benches are generally constructed of iron or steel frames, with 
 boiler-plate tops, and are made wide enough to allow welders to work 
 opposite each other, therefore economising space. The arrangement 
 of hydraulic safety valves is suspended from the main pipe, above 
 the centre of the welding table, making it convenient and handy 
 for the operators. 
 
 The next operation is to fill all the hydraulic safety valves with 
 water until it runs out of the water-level test cock. When this has 
 been done, the rubber tube may be fixed to the outlet tap of the 
 hydraulic safety valve at the one end, and the blowpipe at the 
 other end. Then another piece of tubing is taken and fixed on to 
 the regulator (which is already fixed -into the cylinder valve); 
 the other end of the tubing to be fixed on to the oxygen tap of the 
 blowpipe. All is now ready for the gases for welding. The acety- 
 lene should be turned on at the main cock or tap. After the genera- 
 tor has been charged with a proper quantity of calcium carbide, 
 the holder filled with water, and the purifier charged, all should be 
 ready for the acetylene to be let into the main piping. Pure acety- 
 lene will not come through to the blowpipe on the first starting up 
 of the plant, owing to a quantity of air in the piping and bell, 
 which has to be replaced by the acetylene as soon as generation 
 takes place. Before starting generation, all the taps or cocks 
 fixed in the main pipes should be left open to allow the air to 
 escape as the acetylene starts to come through; as the acetylene 
 has a very pungent smell, it can soon be observed when it begins 
 to come through in quantities. As soon as this period arrives, the 
 taps can all be shut off, and the blowpipe lit. The oxygen being on 
 at the cylinder and regulator, and acetylene turned on at the 
 hydraulic safety valve, and then the two taps on the blowpipe — 
 the oxygen first and the acetylene after the blowpipe is lit — allow 
 it to burn until a good white rigid cone appears. This will take 
 some time, owing to the fact that at first starting up (as already 
 explained) the main piping is partly full of air, although the 
 taps on the main piping have been previously opened. This air 
 mixes with the acetylene, ignites at the tip of the blowpipe, and 
 gives a bluish flame almost like a Bunsen burner. The blowpipe 
 is to remain lighted until the flame reaches a small violet-whitish 
 jet of very clear outline, when it will indicate that the air is all out 
 of the piping, and welding may be done. 
 
CHAPTER XVI 
 METHODS OF WELDING 
 
 The subject to which we are now about to proceed is one which should 
 be very interesting to welding operators. My experience, through- 
 out years of instruction of operators, is that they invariably want 
 to handle a blowpipe before they have even the smallest information 
 of welding or what it means. They soon find out, however, that it- 
 is best to start from the beginning. 
 
 With all appliances in order, and the blowpipe chosen, work can 
 be proceeded with. The pieces to be welded consist of two pieces of 
 angle iron, which are put on the top of the loose bricks on the weld- 
 ing table. The hydraulic safety valve is tested. The acetylene gas 
 is turned on at the tap of the safety valve (the taps on the blowpipes 
 being closed), then the taps on the regulator and the blowpipe are 
 both opened and left open. The cylinder valve is open to let the 
 oxygen through to the blowpipe. The oxygen should be turned 
 off temporarily on the regulator until everything is ready and com- 
 plete for welding, then the blowpipe may be lit, and the flame regu- 
 lated. The blowpipe should be held in the hand, in the central 
 position found by balancing. The weight of the rubber tubes will 
 make the blowpipe feel balanced and light in the hand. 
 
 When commencing welding, the blowpipe should be pointing to 
 the line of welding at a slight angle, so that the blowpipe will melt 
 the metal in advance of the tip of the flame. One must not hold it 
 at too large an angle, otherwise the molten metal will be blown on 
 the cold surface. This point or tip of the flame — that is, the clear 
 white cone — must not touch the metal, but must be about i inch 
 from it. This prevents oxidisation of the metal, and also prevents 
 the nozzle of the blowpipe from being filled up with metal which is 
 blown up with sparks. Further, a much neater weld is made with 
 the flame above the molten mass than in it. 
 
 The blowpipe should be held freely in the hand, and the weld 
 approached at one edge, care being taken that, as soon as the point 
 of melting has been reached, the blowpipe shall be moved slowly for- 
 ward to melt, say, £ inch from the edge. After the melting of this 
 
 83 
 
84 MODERN METHODS OF WELDING 
 
 second part, the blowpipe should be instantly passed over the edge 
 to weld this, which process, from the previous heating, should be 
 almost instantaneous. The object of heating the edge first and 
 not welding, is to stop the molten metal from running from the edge. 
 Nearly all operators, when learning, make this mistake. If, as 
 directed, the second part is made molten and welded, and then the 
 blowpipe is brought over to the end of the weld, this becomes molten, 
 the blowpipe is moved to the third position, the film on the edge 
 of the plate is not broken. Hence it supports the molten of the 
 edge which has been done at the second heating. When proceeding 
 with the welding the blowpipe must be moved forward very slowly, 
 with a gyratory movement. The progress must be regular and con- 
 tinuous, so that the welding may be even and quickly done; and 
 care must be taken that no welding shall be gone over twice, as each 
 time it is done the metal loses some of its niost important constituents 
 and it is therefore burnt and weak. If it is necessary that the two 
 edges of the article that is being welded shall be made molten 
 together, the welding-rod used must also be in the same molten 
 mass together. If the welding-rod is kept in close proximity to the 
 cone of the flame, unity of fusion of both the edges of the weld 
 and the welding-rod will take place, leaving an homogeneous weld. 
 It is best to train the hand, when engaged on small work, to 
 make the flame of the blowpipe describe an elliptical movement, 
 the longer diameter corresponding to double the width of the 
 section of the article being welded. Also, at the same time that this 
 is being performed, it is necessary to combine with this elliptical 
 motion an advancing one ; and, in the advancing, one must keep to 
 the centre line of welding. In the welding of thick plates, where 
 the article is either bevelled on one side or on both, the same ellipti- 
 cal and advancing movement as for light work is required; but a 
 welding-rod is to be used constantly and regularly, so as to fill up 
 the part that has been bevelled. In articles more than § inch thick 
 two layers have to be put on, one after the other, otherwise a sound 
 weld could not be secured. 
 
 The welding-rod is held and directed by the left hand, and should 
 be suspended between the top of the two fingers and the thumb, 
 and over the side of the hand, almost as one holds a pen. It will be 
 found that the rod will thus be balanced nicely for working with 
 accuracy under the tip of the cone. The thickness of the welding-rod 
 should be in proportion to the thickness of the article to be welded. 
 It is very important, as previously stated, that the melting of the 
 feeding-rod and the edges of the weld should take place at the same 
 
METHODS OF WELDING 85 
 
 time, so as to make the edges of the article and the feeding-rod com- 
 bine and form one solid metal. Particular care must be taken to 
 prevent the flowing metal from the rod falling on the unwelded edges 
 of the article. This would cause the weld to be defective. It would 
 be adhesion, not a weld. This, although so simple an error, is often 
 committed, even by experienced welders. More care ought to be 
 taken not to let these simple errors occur. No matter how well 
 
 Butt Joint. 
 
 the other part of the weld is finished, if there is a defect like that of 
 adhesion, it ruins the whole weld. In testing, the least sectional 
 area will be taken. Therefore, if the article is J inch thick, and adhe- 
 sion is carried for J inch deep, the weld is only half as strong. The 
 smallest sectional area is \ inch thick, whereas the article is \ inch 
 thick, being a loss of strength of 50 per cent. Operators should 
 
 Fig. 32. — Indents, not Sufficient Welding-Rod on. 
 
 make these small tests themselves, as this is the quickest and surest 
 method of finding out their own defects. 
 
 The operator should make the rod melt at the same time as the 
 welded edges. The rod is kept in the proximity of the flame, at 
 almost melting-point, so that, as the two edges become melted, the 
 rod is put or dropped on to the molten mass, by passing just through 
 the cone; and, as the welding proceeds at a uniform speed, the rod 
 is progressively worked at a sufficient rate of melting, to fill up the 
 gap in the bevel and to complete the level of the weld to the same 
 thickness as the article being welded. If operators will keep in 
 
86 MODERN METHODS OF WELDING 
 
 mind the information given so far, they should have some idea as to 
 the melting of the article, to the mixing and the forming of the 
 molten mass before solidification, and to the use of the blowpipe. 
 They should be able to grasp the principle of the execution of welds ; 
 but more practice and more study are necessary to make proficient 
 welders. 
 
 Homogeneous welding is the union of bodies by fusion. Dsually 
 the operator does not melt enough, does not get the edges molten 
 before he lets the molten rod fall on the unwelded surface. Or he 
 makes holes, and tries to fill them up by adding the feeding-rod, 
 which usually drops on the cold surfaces. But by quiet determined 
 tuition and practice the student soon comes to know what are 
 the defects and makes great effort to remedy them. He must prac- 
 tise every form of welding, but must first of all master the commonest 
 joints, such as the butt and the edge. These joints must be done 
 over and over again before he is allowed to proceed with further 
 kinds of welds. This is imperative if he intends to become a first- 
 class operator. Two joints should be made : first, weld with welding- 
 rod ; second, without welding -rod ; third, with welding -rod, hammer, 
 and anneal. 
 
 The two tests should be butt and edge, and ^-inch thick steel. 
 "Bending test" may be made by fixing in the vice the article 
 welded with the line of the weld just | inch above the vice, then 
 bend right over with a hammer. "Area test " : Two pieces put at 
 right angles, and weld along the corner, seam, then flatten out with 
 hammer ; see if the area is the same thickness as the material welded. 
 Third test: Two pieces ^ inch thick; butt and weld; use plenty of 
 welding-rod. After welding, heat the piece up to white heat by 
 blowpipe, hammer to thickness of metal ; repeat the heating by the 
 blowpipe, and allow to cool; and then bend in the vice. Watch 
 the difference between the annealed one and the one not annealed. 
 If these tests fracture when bent, repeat the welding and testing of 
 these two articles until they can be done without cracking or fracture. 
 
 Operators in the first few courses of training generally point the 
 flame on to one of the faces of the bevel. As the metal melts it 
 runs down on the other " cold " side. It is usually covered over 
 by the molten metal, and is not seen externally; but the defect is 
 there, nevertheless. Operators must not allow this to occur, but 
 must remember that the simultaneous and regular melting of the 
 two edges of the weld and the welding-rod is a very important point. 
 If good welds are to be obtained, this is a sine qua non. The begin- 
 ning of the weld is always slower, and the end more rapid, because the 
 
METHODS OF WELDING 
 
 87 
 
 temperature of the article is increased as the line of welding reaches 
 the end. The finishing must be done sharply, otherwise the metal 
 gets so molten that the ends give way and allow the molten metal 
 to run away. 
 
 We may now proceed to autogenous problems. The vast amount 
 of welding carried through during the war was really amazing. The 
 thousands engaged on this process were chiefly concerned with 
 mild steel articles for military purposes. Millions of feet of sheet 
 
 Fig. 33.— Round Bar Bevel, Chisel Points. 
 
 steel were welded, and the task of training these temporary operators 
 was huge. Of course, there was no time to impart a thorough know- 
 ledge of welding. They were given just sufficient instruction to enable 
 them to get along and practise for themselves. It was remarkable, 
 however, how quickly they settled down to the plain welding of mild 
 steel articles. These welds apparently are the most easy to obtain, 
 but in reality, to do them as they ought to be done, so that they will 
 
 M 
 
 Fig. 34. — Tube Bevel. 
 
 stand any test applied, is not so easy. Indeed, they require on 
 the part of the operator the utmost thought and care. It is 
 well known that 90 per cent, of welding is in mild steel, therefore 
 the operator will have to be taught the constituents and analysis 
 of these metals. He must also make himself acquainted with the 
 melting-points of the metals and their oxides. He must study the 
 articles on the different metals in other parts of this book, and must 
 especially learn to know the melting-points of all metals and their 
 oxides. This is most important in the cases where the metals and 
 their oxides differ in their melting-points. Take, for instance, steel 
 
88 
 
 MODERN METHODS OF WELDING 
 
 and aluminium. Steel is 1,600° C, aluminium 700°' C. In the case 
 of the oxides the difference is very much greater. Aluminium melts 
 at 700° C, but its oxide melts at 1,600° C. ; this is why operators 
 do not find aluminium welding easy of execution. The metal itself 
 readily forms in globules or beads, which do not run together 
 owing to the outside film not being melted. These films are the 
 oxide. The same happens to all metals, but others are not so much 
 defined as aluminium. 
 
 There are many defects that happen in welding which could be 
 avoided if operators would see if they were corrected as they went 
 along. Some of the defects in welding mild steel will be described 
 below. 
 
 The first is the lack of penetration — that is, either the power of 
 the blowpipe has not been large enough or the operator has passed 
 
 Fig. 35. — Fracture after Bending. Fig. 36. — Lack oe Penetration. 
 
 over the line of welding far too fast. This lack of penetration is a 
 serious defect, and is a defect that is commoner than any other, 
 especially in the case of butt welding. By not penetrating right 
 through the metal, the original thickness of the metal is reduced 
 in sectional area, because the weld has only penetrated to two-thirds 
 of its thickness. This is very serious, and in the case of tubing which 
 has to stand pressure very often the weld gives before the full 
 pressure is put on. This non-penetration occurs sometimes when 
 the edges are not bevelled. The heat has not been sufficient for 
 the fusion to go right through the whole thickness of the joint, 
 and also the thickness not welded constitutes a starting-point for 
 a break. 
 
 Figs. 35 and 36 show the weld, and the result of the bending. 
 
 On the other hand, operators should take care not to penetrate 
 too far through, or the metal runs through from the molten bath 
 above, and often leaves a conical hole which is difficult to fill up and, 
 in doing so, generally oxidises the metal owing to being too long on 
 the weld, and the metal is burnt. 
 
METHODS OF WELDING 
 
 Adhesion is another defect which operators often make, especi- 
 ally in thick bevelled work, owing either to not having a powerful 
 enough blowpipe or to too heavy a pressure of oxygen, which 
 " swills " the surface of the bevel, without taking time to make it 
 liquid; dropping the liquid rod on this "swilled" surface causes 
 it to stick or adhere. This is not fusion. Then, again, some opera- 
 tors do not melt both edges together, and therefore, again, no fusion 
 or unity takes place. Again, many welds are interposed with oxide. 
 This arises from several causes. The metal may not have been 
 properly liquefied, which causes blowholes to form in the interior 
 of the molten mass and remain after solidification. Likewise, if the 
 metal gets too hot and boils, this causes gas to be imprisoned in the 
 interior of the weld, also producing blow-holes. The defect may 
 also arise from not attacking the bevelled edges of the weld suffi- 
 
 Fig. 37. — Single Bevelled Joint. 
 
 Fig. 38. — Adhesion, Bad Weld. 
 
 ciently, or attacking them unequally, or from the flowing of the 
 metal or the liquid welding-rod on the edges that have not been 
 melted. 
 
 Operators must pay attention to this point and understand that 
 the molten metal flowing from the edge of the weld brings about 
 adhesion if this part itself is not melted. It is very important 
 indeed that, before the bevelled edges are melted, the bottom of the 
 bevel should be melted first, so that as the sides of the bevel are 
 melting, the liquid metal runs into the molten bath lying ready to 
 receive it at the bottom of the bevel. If this is carried out there 
 will be no adhesion. 
 
 It is important that all welds, no matter for what job, should 
 always be well filled. There must be no part of the weld, on either 
 side, under the full section of the metal which is being welded. 
 If it is, the article loses its strength. This is almost as much to be 
 avoided as a bad weld. Assume a tube is being welded j inch thick. 
 A butt joint is being made, and it is welded along the line of the 
 
90 
 
 MODERN METHODS OF WELDING 
 
 weld with what is, from the operator's point of view, a good weld. 
 It is inspected and found to have, in two or three places along the 
 outside lines, indents about ^V inch deep. These indents were caused 
 by the melting of the original metal, which flowed with the circular 
 movement of the blowpipe, and left the indents behind. These 
 should have been filled up with the welding-rod as the welding pro- 
 ceeded. Consequently, through this error, the tubing will not stand 
 the full amount of test designed for J inch thick. The finished 
 sectional area is only -$■% inch thick. If ^ would do, why use \ inch 
 thick ? Operators, however, must learn to keep the welds up to 
 the thickness of articles. These articles are skilfully designed and 
 the stresses all calculated out for the purposes to which they 
 are to be put; therefore, if the welding all over is not up to that 
 
 First Start 
 oP Fracture 
 
 Fig. 39. — Not Fully Penetrated, 
 First Starting of Fracture. 
 
 Fig. 40. — Not Penetrated Through. 
 
 particular thickness the article is not so strong. In some cases 
 this is important. 
 
 Tests for Welds. — The majority of defects are hidden in the body 
 of the weld, and the operator is often ignorant of them. But there 
 are many ways of testing welds, so that after he finds them out 
 he should be able to overcome all his defects. This will take some 
 time, but with patience and close study he should be successful. 
 One cannot, of course, break joints of commercial work in order to 
 get stresses and strains or examine the internal constitution. But 
 with facilities to make his own test-pieces of similar metal to those 
 he has worked with, the operator is able to make all the tests neces- 
 sary, including resistance and elongation. Test by corrosion, or, 
 as it is called, the micrographic test, may be applied to all metals 
 of yV mcn or over. 
 
 Two pieces of flat bar should be welded, about 3 inches long, 
 butt- jointed, and then cut through the longitudinal way, across the 
 weld. One of the faces is to be polished to a spotless surface, all 
 
METHODS OF WELDING 
 
 91 
 
 grease is removed, and the etching fluid, applied with a brush, 
 soon exposes any defect, adhesion, or oxide. A plain black image 
 appears which shows all flaws very plainly, and if it has been burnt 
 this also can be clearly seen. It is important that the face be polished 
 well and free from grease. The polishing does not disclose any 
 defect, unless it be a bad one ; but when the etching fluid is applied, 
 the defects appear instantly. Therefore, if operators will from time 
 to time practise these tests, they will soon remedy the defects. 
 
 Fig. 41. — Bending a Test-Piece in a Vice. 
 
 Etching liquid may be as follows : 
 
 Iron and steel 
 
 [water . . . . 10 parts. 
 
 Iodine solution^ potassium iodine 2 ,, 
 
 [iodine . . . . 1 part. 
 
 The solution is applied with a brush immediately the polished 
 cut is ready. The structure develops almost suddenly, and in a 
 few minutes the corrosion is sufficient. Wash with running water, 
 dry with alcohol, and cover with a layer of varnish if the test is to 
 be preserved. The test by bending is one applied in most technical 
 schools, and is appropriate for all ductile metals. A test-piece may 
 be made out of f-inch diameter round steel. This must be first 
 prepared for welding by shaping two ends to chisel type ends. 
 The reason why it must be pointed as a chisel is that, as the metal 
 is welded, it falls down to the bottom where the chisel-point meets, 
 at the centre of the round bar. If it were not for the chisel-point, the 
 
92 MODERN METHODS OF WELDING 
 
 article being round, the metal would run to the bottom of the 
 welding table and spread about instead of building up. The welding 
 of this round piece should be executed and finished off the same 
 thickness as the bar itself, so as to give it a fair test when bending; 
 after welding and cooling it can be put in the vice, with the weld 
 just over the top of the jaws, the hammer applied in any direction. 
 It should be bent over to a radius of 1J inches. If this is done 
 without a crack or fracture, the work will pass. 
 
 The hammering and annealing test is the best of all. Operators 
 should weld several of these small test-pieces in various sections of 
 steel. One-third of the test-pieces to be put through the three 
 tests of corrosion — bending, hammering, and annealing — should be 
 pieces where welding only has been done ; one-third should be pieces 
 which have been welded, annealed, and- hammered; annealed and 
 cooled slowly. A record of all these tests should be kept, and the 
 result will be surprising. The illustration (Fig. 41) shows a test- 
 piece in a vice; this is a very easy method of the bending test; 
 students who can execute welds to stand this test should be able to 
 tackle all ordinary commercial work. 
 
CHAPTER XVII 
 PREPARATION OF WELDS 
 
 It is impossible to overestimate the importance of thorough pre- 
 paration of the work before the weld is actually attempted. Any 
 time spent in this way is amply repaid afterwards in the easier 
 execution thus made possible. The preparation varies considerably 
 with the nature of the metal and the thickness and form and posi- 
 tion of the parts or articles to be welded. It is impossible to lay 
 down any hard-and-fast rules. For each metal we may have to 
 adopt a different preparatory procedure. For instance, some of 
 the metals have much lower melting-points than others, and must 
 be dealt with accordingly. 
 
 The general principles obtaining in the best practice direct 
 that the line of weld must be opened out — -that is, the two sides 
 bevelled, each to 45 degrees, making an angle of 90 degrees. This 
 is to make certain that the weld shall be penetrated, and not merely 
 sealed over. The welding must be done from the bottom of the 
 bevel and properly filled in with the welding-rod, the metal at the 
 edges being molten at the same time, so as to unite with the molten 
 rod; the two combining make a good sound weld. Bevelling also 
 increases the area of the surface of the weld, thereby strengthening 
 the latter. It allows the addition of a larger quantity of metal of 
 better quality, since rods of pure iron are added to the welds. 
 Bevelling is carried out on thicknesses of £ inch. After it 
 reaches \ inch and over in thickness, the bevel should be on both 
 sides. The illustrations on p 94 show two pieces. 
 
 It is essential that the line of welding should be thoroughly 
 cleaned (particularly in the case of aluminium), either by hand 
 tools or by some chemical agent. Too much stress cannot be put 
 on this, as welders often find. The most important part of the pre- 
 paration is that of arranging the pieces to be welded in such a posi- 
 tion that there will be no deformation, fractures, cracks, or internal 
 strains, and that they will be linable at the conclusion of the opera- 
 tion. This applies chiefly to cast-iron articles, which are generally 
 
 93 
 
94 MODERN METHODS OF WELDING 
 
 intricate castings of various thicknesses and irregular shapes and 
 are often cumbersome. 
 
 This is a point in which the skill of the operator is revealed, 
 as there are no fixed rules to guide him. Welders too often fail 
 to take sufficient precautions to keep the article adjusted and in a 
 correct position for welding, so as to allow for that phenomenon 
 known as expansion and contraction, and to leave the welded article 
 finable at the finish. One must keep in mind that old maxim: 
 
 Flat Bab Double Bevel. 
 
 " What is worth doing is worth doing well." A weld well prepared 
 is half done. As the result in welding depends to a certain degree 
 on how the preparation has been carried out, one cannot spend too 
 much on it, especially where intricate, uneven castings are concerned. 
 It is the interested, careful, and thoughtful workman who gets success 
 in the welding of such articles. 
 
 Bevelling should be done on both sides, as has been said, if over 
 
 Fig. 43. — Angle Iron, Bevelled One Side. 
 
 \ inch thick. If it cannot be got from both sides, then a deeper 
 bevel must be made on the one side. Operators sometimes do not 
 bother about bevelling, even if it is a case of \ inch thick without 
 bevelling. To omit it always leads to bad results — such as bad 
 penetration, adhesion, or overheating of the metal — and usually 
 leaves about \ inch unwelded at the bottom edge. This reduces 
 the sestional area, and is the means of starting a fracture (see 
 the illustrations below), which is what happens when the articles 
 are not welded through. If the above defective weld were tested, 
 it would hardly stand a tensile test of three-fifths of the original 
 
PREPARATION OF WELDS 
 
 95 
 
 strength of the metal. Taking a good view of the above section, 
 one sees that the weld does not touch the bottom joint of the bar. 
 Further, it can be clearly observed that the molten metal did not 
 penetrate through the full thickness, but formed itself into a semi- 
 circular mass at the bottom joint of the bar. This semicircular 
 line reduces very much the area of the weld, and with it the strength. 
 If a proper weld had been made, a f-inch plate would have been quite 
 
 Fig. 44. — Not Penetrated. 
 
 as strong, and a great saving of material in the thickness of the 
 respective plates would have been effected. In the illustrations the 
 lines marked A and B show the thickness of the metal not welded. 
 This is a great consideration from an engineer's point of view. He 
 designs work calculated, on a specified thickness, to stand certain 
 stresses; and it may, in any particular instance, be a very important 
 
 Fig. 45.- 
 
 -Sface between A and B shows Amount Unwelded, and Fracture 
 when Bent. 
 
 job, where the stresses must be kept up to counteract the design. 
 But if a bad weld is made and does not penetrate to the thickness, 
 then the article will fail, and cause heavy loss. On the other hand, 
 if one was sure of always penetrating the weld (which can only be 
 done by preparing and bevelling), articles can be designed with 
 lighter materials, thus saving expense. 
 
 The above sketches show very clearly how defects occur; the 
 welding has not been penetrated through the whole thickness. 
 
96 
 
 MODERN METHODS OF WELDING 
 
 hence the strength is not up to the thickness of the original metal. 
 Upon bending the bar, the unweldecl line opened, and also started a 
 fracture along the line of welding. 
 
 One cannot be too emphatic in stating that pure metals, pure 
 welding-rods, pure gases, are most essential to good welds. Adjust- 
 ment before welding is a point in which the experience and skill of 
 the operator tells. There are very few rules for his guidance, as 
 the articles are so different that each one has its own particular 
 scheme. Upon any work but that of the simplest character, failure 
 to grasp and apply the laws of expansion means partial or total rum 
 of the work. It is impossible to control expansion and contraction 
 
 Fig. 46.- 
 
 -Machine Operator Welding Seams 116 Inches Long in Corrugated 
 Sheets for Transformers. 
 
 by physical or mechanical forces. The only way to prevent disastrous 
 results is to foresee the probable direction and extent of the pheno- 
 mena and nullify the effect by preheating the whole or certain parts 
 of the work, either by the blowpipe or the welder's furnace, the latter 
 being recommended. 
 
 Operators must take precautions with all articles of non-ductile 
 matals, to see that they are all properly adjusted, and carefully 
 fixed at some part of them, to prevent them being moved or dis- 
 turbed during the process of welding. They must provide that ex- 
 pansion and contraction shall take place uniformly, so that the 
 
PREPARATION OF WELDS 
 
 97 
 
 article is not, after welding, distorted in any way nor out of 
 alinement. There are various ways of making these fixtures, and 
 a series of different sizes should be kept. One is illustrated below, 
 which is easily manipulated, even when hot from the furnace. 
 
 Expansion in physics is an enlargement or increase in the bulk 
 of bodies, in consequence of a change of their temperature. This 
 is one of the most general effects of heat, being common to all bodies 
 whatsoever,, whether solid or fluid. The expansion of solid bodies 
 is determined by the pyrometer. One can realise the force of expan- 
 sion from water that has frozen in an enclosed vessel or pipe. When 
 the careless motorist leaves the cooling water of the engine in the 
 water-jacket of the cylinder overnight, on a frosty day, the next 
 
 Fig. 47. — Casting Broken. 
 Left: Bevelled and cramped setting for welding. Right: Welding completed. 
 
 morning there may be ice. On which heating by water or other- 
 wise, the ice expanded, and this powerful force fractures the cylinder 
 water-jacket owing to there being no outlet for the expansion. 
 
 This same phenomenon is seen with all cast-iron articles. It 
 is useless to attempt by force to oppose this expansion and contrac- 
 tion. The method is to avoid or limit the consequences. If not, 
 any welding done on non-ductile articles will surely cause deforma- 
 tion, cracks, and internal strains. The whole articles should be raised 
 to a temperature, in the case of cast iron, of not less than 900° C. 
 and not exceeding 1,100° C, and then quickly welded, and returned 
 to the annealing furnace to cool slowly. 
 
 It must be understood that there is no other means of obtaining 
 good results on non-ductile articles, except that of preheating and 
 annealing on properly designed and well-thought-out lines. In cases 
 
98 
 
 MODERN METHODS OF WELDING 
 
 45" Bevel 
 
 ~=§*& 
 
 Finished weld. 
 
 = jFT -i 
 
 g j 
 
 / r /^. 4$. Proper set up for butt 
 welding of pipe, and finished 
 weld. 
 
 Bevel here 
 
 Section showing 
 finished weld. / 
 
 Fig. 49- Welded cross in pipe, left , 
 bevel right, welded. 
 
 Bevel here 
 \ 
 
 Section showing 
 finished weld. 
 
 F>9- 
 
 \/=== 
 
 50. Showing construction of 
 
 T pipe, bevel and weld 
 Bevel here. 
 
 Section showing 
 finished weld 
 
 Fig. 51. Showing pipe at 45 degrees, 
 weld and bevel. 
 
 Figs. 48 to 51. 
 
PREPARATION OF WELDS 99 
 
 where the articles are free to expand and contract and are fairly even 
 in thickness and size, welding may be done without preheating. 
 These articles will not crack when welding; but they are few, and 
 should be carefully watched. After welding they should be placed 
 in the annealing oven to remove all internal strains, and left to cool 
 slowly. 
 
 The illustration on p. 97 is a case where it is not necessary to 
 take expansion into consideration, as there are no tied ends. The 
 casting may therefore be welded from the cold state ; but, from an 
 economical point of view, it should be heated to save the gas. The 
 illustration referred to shows how a casting should be prepared for 
 welding. The line of bevelling can be seen; the under side is 
 similarly bevelled ; the fixing of the cramps will be noted, holding 
 the casting in position while the welding is performed. 
 
 In making the set-up for butt-welding pipes the edges should be 
 separated sufficiently to allow the heat of the welding flame to drive 
 all the way to the bottom of the weld. This separation, however, 
 should not exceed ^ inch, because too great separation is conducive 
 to the formation of large bumps of metal within the pipes, which is 
 very undesirable. Many operators butt the edges of the pipe 
 " square up " and do not attempt to secure complete penetration. 
 The weld is slightly reinforced and is virtually as strong as any portion 
 of the pipe. The edges of the butt weld in the pipe and, more par- 
 ticularly, those of other fittings shown herewith should be bevelled 
 Simple as these lay-outs appear, the average operator experiences 
 more or less difficulty in getting the various parts to line up satis- 
 factorily. 
 
 Figs. 48 to 51 show several tube joints, described above. 
 
CHAPTER XVIII 
 WELDING TABLES 
 
 Small welding tables are used in both small and large workshops. 
 There is quite a variety of types, from which one may select any 
 particular one. They are built so as to give an easy position to the 
 operator, who can work all round it. 
 
 This facilitates the task very much, and the weld is done quicker 
 and better. The tables are portable and can be moved anywhere. 
 
 Fig. 52. — Mild Steel Operator's Table. 
 
 They are usually constructed of angle-iron formation, and all the 
 joints are welded. They may be light, and 1J by ■§• angle steel is 
 strong enough for all ordinary purposes. The making is quite 
 simple, being a matter of a few hours for any intelligent operator. 
 Firstly, there are two frames, say, 2' 9" X 1' 9", which can be 
 made in one piece; cut a V out of each corner; then bend at those 
 places where cut out and weld up. Then the four legs should be 
 welded to the frames, the one of which is on the top of, and the other 
 inside, the legs, 8 inches from the bottom. This bottom frame should 
 
 100 
 
WELDING TABLES 101 
 
 be fitted with a mild steel plate, dropped closely into the angle frame, 
 forming a tool table, where the operator can keep his tools and weld- 
 ing-rods. Tables are usually made 2 feet 3 inches high, but this may 
 be altered to any height. 
 
 The following are illustrations of ordinary types, but they may 
 be varied to suit circumstances. 
 
 One of them is shown plain and one with fire-bricks. Notice 
 strips welded across the top frame to carry the bricks; the other is 
 shown with fire-bricks in position. It is very necessary for fire- 
 bricks to be put under articles to be welded, because the brick retains 
 
 Fig. 53. — Adjustable Welding Table, with Vice Attached. 
 
 the heat, which assists the heating of the articles being welded, 
 thereby saving gas. This is much better than having a bare steel- 
 plate table, which often warps by the continual heating. The size 
 of the top may be determined by the layer of bricks, and these should 
 be laid out before the frame of the table is made, so as to get the 
 correct size for the bricks to fit well together. 
 
 Fig. 53 is an adjustable type and can be turned at any angle. 
 This is found exceptionally useful on repairs, when the welding 
 is not horizontal. One has a vice attachment, and is very useful. 
 
 In large shops, where there are a large number of operators on 
 repetition work, it is a practice to construct long tables at which 
 thirty to forty operators can all work. They usually work face to 
 
102 MODERN METHODS OF WELDING 
 
 
 e i' 1 " a 'i.< 
 
 Fig. 54. — Welding Table with Fire-Bricks. 
 
 Main from Generator 
 
 ^ 
 
 U 
 
 ^ 
 
 ^ 
 
 Fig. 55. — Table for a Number of Operators. 
 
WELDING TABLES 
 
 103 
 
 face — that is, in pairs each side of the table. The acetylene main 
 gas pipe is usually brought over the centre of the table, and the 
 hydraulic safety valves are suspended from the main supply to a 
 convenient position for the operators. They are spaced according 
 to the requirements of the articles being welded without crowding. 
 
 Fig. 56. — Tilting Table. 
 Articles can be bolted on. 
 
 In construction, these tables are almost identical with the smaller 
 ones, with the exception that no frames are used, except the framing 
 under the boiler-plate top. Such large tables usually have large 
 flat bricks, about 1J inches thick. Fig. 55 is a sketch of one, 
 showing the position of the acetylene main gas-supply, from which 
 are suspended the hydraulic safety valves, one for each operator. 
 The tables may be extended to any length, to the limit of the shop 
 and the supply of acetylene. 
 
CHAPTER XIX 
 FURNACES FOR HEATING 
 
 In the repair of non-ductiles, it is necessary to heat them up before 
 the welding can take place. It is impossible to make satisfactory 
 welds of cast-iron or aluminium articles without first preheating to 
 a level temperature. If an attempt is made to weld any articles of 
 the metals stated above, without getting hot all over, fractures take 
 place. For instance, take a casting of iron just as it is, without 
 preheating. Apply the blowpipe at any place, get a good heat, 
 and the result is that you will hear it crack in some part other than 
 where heated. When the casting gets cold again, further fractures 
 are sure to occur. These are caused by the uneven heating of the 
 
 Fig. 57. — Kerosene Preheating Torch. 
 
 article in one place with the blowpipe, causing the non-conducting 
 metal to expand where it was hot, while the cold part of the casting 
 remained normal. 
 
 The methods adopted to overcome this expansion and contrac- 
 tion on non-ductile articles are very simple, if care is taken to pro- 
 vide the right appliances. One is to use a furnace with coal, coke, 
 charcoal, or gas. Coal would be all right in a properly constructed 
 furnace, where the flame is reverberatory, and a second floor has 
 been made for the heating of the articles which are not in actual 
 contact with the fire. But the author's experience is that this 
 method is much too costly to maintain. The expense would be 
 
 104 
 
FURNACES FOR HEATING 
 
 105 
 
 prohibitive in a small shop, unless there was a good quantity of 
 castings to weld daily. Also the heat is too fierce, and would not 
 do at all for aluminium, which would probably collapse before 
 it could be got out. The coal furnace should only be used in large 
 shops. Although there are many of these furnaces being used, 
 with coal, coke, and charcoal, and, to a certain degree, giving 
 
 d5=d«^^£ 
 
 Fig. 58. — Gas-Heated Preheating Oven. 
 
 satisfaction, they are generally employed in conjunction with other 
 work. The best method, with the least initial outlay, the most 
 economical working, and the best even heating, is a furnace heated 
 by gas at either ordinary or high pressure. Fig. 58 shoAvs an 
 excellent type, which gives exceptionally good results; it is not 
 hard to make, and it will not be expensive. 
 
 Fig. 57 shows a preheating kerosene torch. This is ideal for 
 
106 
 
 MODERN METHODS OF WELDING 
 
 the repair shop; it is light, portable, and owing to a patented 
 sliding valve, can be operated in any position. It produces a 
 clean even flame of about 3,000° F. and 24 inches long. It con- 
 sumes about 1 gallon of kerosene per hour. The tank capacity 
 is 3 gallons. The burner thoroughly vaporises the kerosene, and 
 the flame cannot blow out in the wind. It is very good for pre- 
 heating work and also for annealing. The weight is 25 pounds. 
 
 The preheating oven or furnace shown on p. 105 is easy to con- 
 struct from sheet steel. It was designed by the author, and has 
 proved very efficient in use. It may be made any size to suit the 
 work in hand. The one illustrated is 4 feet deep, 3i feet wide, 3 feet 
 long. The construction consists of an inner and outer casing, with 
 
 M 
 
 u^ 
 
 Fig. 59. — Folding Asbestos Screen, eoe Preheating Furnace. 
 
 1J inches space between them, filled with asbestos. The back of 
 the stove (inner and outer) must have the asbestos put between 
 before bolting up the stove. There is an outlet at the back, near 
 the top, to allow the burnt gases to escape, and the piping should be 
 carried from the outlet to the atmosphere. On the inside of the 
 stove are bearers or ledges for perforated trays, in which trays the 
 articles for heating are fixed. The door is also made of two sheets 
 of steel, with asbestos between them, and fits very closely to prevent 
 any cold air from getting into the interior. Along the bottom are 
 four Bunsen burners. These may be purchased from any reliable 
 firm, but they must be the best that can be got. 
 
 A great saving in gas can be made by adopting, for the purpose 
 of extracting all the heat from gas, a folding asbestos cover as in 
 Fig. 59. It is made like a fourfold screen. After the article to be 
 
FURNACES FOR HEATING 
 
 107 
 
 heated is put in the stove, and before the gases are lit, it is placed 
 on the top of the article. When the gases are lit, this cover concen- 
 trates the whole of the heat, thereby getting the article heated in a 
 much shorter time, and saving quite 50 per cent, of the gas. This 
 asbestos cover can be used for preheating and also for annealing. 
 The screen described may be made in smaller or larger sizes to suit 
 the work required. The perforated tray in the furnace is made to 
 draw out with the article on it. 
 
 The handling of castings, such as a four-cylinder motor casting, 
 when hot, is a very troublesome job if there are no proper appliances. 
 Hence, fractures often occur through not getting the articles quickry 
 enough into the annealing furnace, and allowing the temperature 
 to fall into the expansion zone — that is, 850° C. The author has 
 
 Fig. 60. — Light Lifting Crane for Use in Lifting Castings In and Out 
 of the Preheating Furnace. 
 
 known this to occur on many occasions. Good welds have been 
 executed, but they have been left one or two minutes too long before 
 getting them into the annealing furnace, and very often the fracture 
 is much larger than the original fracture. An appliance which will 
 be found to overcome this difficulty of removing and inserting the 
 hot articles rapidly, with ease and comfort, is shown in Fig. 60. 
 This should be installed in all workshops where they are dealing 
 with heavy repairs, which have to be preheated. It is simply a 
 small swinging overhead crane, very lightly built, to carry about 
 5 cwt. It has two small wheels, running along a flat bar on edge. 
 Suspended from the plates of the wheels is a lever about 9 feet 
 long, a hook on the end, and a rope at the other. 
 
 The rope is for lifting the weight suspended at the other end ; on 
 the hook may be hung a two- or four-legged chain to lift the articles 
 
108 MODERN METHODS OF WELDING 
 
 already fixed on the oven tray. The hot casting may thus be put 
 in and out of the oven or furnace with ease, and without any of the 
 welders getting burnt. 
 
 The phenomenon of expansion and contraction with reference 
 to metals is one of the most important problems in the welding pro- 
 cesses. No operator can become competent without devoting long 
 study to it. Cast iron is a non-ductile metal — that is, it can only 
 expand and contract in its whole piece. Hence, if heat is applied to 
 one part of the article, the heated part becomes expanded past its 
 elastic limit. The cold part of the article does not expand and the 
 heated part is therefore fractured. On the other hand, if the whole 
 article had been heated, by putting in a- proper heating furnace, 
 and the heat carefully regulated until it had reached a minimum of 
 900° C, the article could have then been taken from the furnace and 
 put on the welding table, welded, and put back into the annealing 
 furnace and allowed to cool slowly, preventing any cold air from 
 getting near it, when the article would be quite sound without 
 crack or fracture. 
 
 This operation of expansion and contraction is so important, and 
 the methods to counteract it are so easy, sure, and effective, 
 that, in all cases where welding of non-ductile articles is done, one 
 must see to it that these operations of preheating and annealing are 
 carried out. 
 
CHAPTER XX 
 IRON AND STEEL 
 
 The mechanical properties of metals are often, in a great measure, 
 dependent on the thermal treatment to which they have been sub- 
 jected. There can be no question that the application of heat to a 
 metal may produce a remarkable molecular change in its structure, 
 the nature of the change depending on that of the metal, and on the 
 treatment it has undergone. It will be well, therefore, to consider 
 carefully what happens when the metals are submitted to three 
 principal operations involving thermal treatment, which are known 
 respectively as annealing, hardening, and tempering. Usually all 
 three are intimately related. Annealing may be defined as the 
 release of the strain in metals, which may itself have been produced 
 by mechanical treatment, such as hammering, rolling, or wire- 
 drawing, or by either rapid or slow cooling from a more or less 
 elevated temperature. 
 
 As an example of the former, it may be mentioned that metals and 
 alloys which have been rendered excessively hard by rolling are 
 heated usually to bright redness and allowed to cool slowly. In the 
 case of copper it does not appear to be important whether the cooling 
 is slow or rapid, and in recent years much experimental evidence 
 has been accumulated which tends to show that, in the case of certain 
 metals which have been hardened, a more or less prolonged exposure 
 to a low temperature under 100° C. will sensibly anneal them. 
 On the other hand, the rapidity with which cooling is effected is very 
 important. Bronze containing about 20 per cent, of tin is rendered 
 very malleable by rapid cooling. In the case of iron and steel the 
 thermal treatment is especially important. 
 
 Steel, it must be remembered, is modified iron. The name iron 
 is, in fact, a comprehensive one, for the mechanical behaviour of 
 the metal is so singularly changed by influences acting from within 
 and without its mass as to lead many to think that iron and steel 
 must be two distinct metals: their properties are so different. 
 Pure iron may be prepared in a form as pliable and soft as copper 
 — -for instance, the charcoal used for welding. Steel can readily be 
 made sufficiently hard to scratch glass. Notwithstanding this extra - 
 
 109 
 
110 MODERN METHODS OF WELDING 
 
 ordinary variation in the physical properties of iron and certain kinds 
 of steel, the chemical difference between them is small, and would 
 hardly secure attention if it were not for the importance of the 
 results to which it gives rise. 
 
 It is necessary to consider the nature of the transformations 
 which iron can sustain, and to see how it differs from steel. Its most 
 useful and advantageous property is that of becoming extremely 
 hard when ignited and plunged in cold water, the hardness produced 
 being greater in proportion as the steel is hotter and the water colder. 
 The colours which appear on the surface of steel slowly heated 
 direct the artist in tempering or reducing the hardness of steel to 
 any determinate standard. 
 
 Hardening is the result of rapidly cooling a strongly heated mass 
 of steel. 
 
 Tempering consists of reheating the hardened steel to a tempera- 
 ture far short of that to which it was raised before hardened. This 
 heating may or may not be followed by rapid cooling. 
 
 Annealing, as applied to steel, consists in heating the mass to a 
 temperature higher than that used for tempering, and allowing it to 
 cool slowly. 
 
 This may be seen experimentally in the following manner : Three 
 strips of steel of identical quality may be taken. It can be shown 
 by bending one that it is soft, but if it is heated to redness and 
 plunged in cold water it will become hard and will break on any 
 attempt to bend it. The second strip may, after heating and rapid 
 cooling, be again heated to about the melting-point of lead, when it 
 will bend readily, and will spring back to a straight line when the 
 bending force is removed. The third piece may be softened by being 
 cooled slowly from a bright-red heat. This will bend easily and 
 will remain distorted. 
 
 The metal has been singularly altered in its properties by com- 
 paratively simple treatment. And all these changes, it must be 
 remembered, have been produced in a solid metal, to which nothing 
 has been added, and from which no material has been taken away. 
 The theory of the operation described above has been laboriously 
 built up ; its consideration introduces many questions of great interest 
 both in the history of science and our knowledge of molecular physics. 
 
 Physical Properties of Metals. 
 
 Molecular Structure. — The physical aspects of metal are so pro- 
 nounced as to render it difficult to abandon the old view that metals 
 are sharply defined from other elements, and form a class by them- 
 
IRON AND STEEL 111 
 
 selves. Like all other elements, metals are composed of atoms 
 grouped in molecules. Any force that alters the relations of the 
 atoms in molecules modifies the physical properties of the metals. 
 Indeed, it would be easy to show that the physical constants of each 
 metal vary with its degree of purity. The molecular grouping of 
 metals is doubtless very varied, and little definite is known 
 regarding the structural stability of most of them; but it may be 
 assumed that it is not very great, as some metals split up into single 
 atoms when they are volatilised, and most of them unite readily 
 with chlorine and oxygen. Consequently, any mass of which the 
 fundamental molecules are the constituent particles may practically 
 be regarded as a single molecule. Two fundamental molecules must, 
 however, be held to be capable of uniting to form complexes that 
 have less power of cohering, and any circumstances tending to bring 
 about the formation of such complexes would tend to make the 
 material less tough. This may account for the extraordinary altera- 
 tion in the properties of many metals produced by very small quan- 
 tities of incompatible foreign matters. 
 
 Density. — The density of the metal is dependent on the intimacy 
 of the contact between the molecules. It is dependent, therefore, 
 on the crystalline structure, and is influenced by the temperature 
 of the casting, by the rate of cooling, by mechanical treatment, by 
 the purity of the metal. All metals, except bismuth, are lighter 
 when molten than in the solid state. In the case of cast iron, which 
 passes through a pasty state on solidification, the density is aug- 
 mented by wire-drawing, hammering, and any other physical method 
 of treatment in which a compressing stress is employed. Mere 
 traction, however, may diminish the density by tending to develop 
 cavities in the metal. Pressure on all sides of a piece of metal 
 increases its density. A metal can only be compressed if the 
 result of the application of pressure is to cause it to pass to an 
 allotropic state — that is, denser than that which it originally 
 possessed. 
 
 Fracture. — The appearance of the fractured surface of a metal 
 depends partly on the nature of the metal and partly on the manner 
 in which solidification occurred. Sudden cooling, to a great extent, 
 prevents the formation of crystals, while slow cooling facilitates 
 their development. Long, continual hammering, frequent vibra- 
 tions, and intense cold will produce the latter result. Any condition 
 that affects either the cohesion or the crystalline structure of a metal 
 affects its fracture. 
 
 Malleability. — This is a property of permanently extending in 
 
112 MODERN METHODS OP WELDING 
 
 all directions, without rupture, through pressure produced by slow 
 stress or by impact. As a rule, crystalline metals are not malleable, 
 and any circumstances that tend to produce crystallisation must 
 affect the malleability. Thus, in nearly all metals, the malleability 
 becomes impaired when they are subjected to rolling or long- con- 
 tinued hammering. But this property may be regained by anneal- 
 ing, which consists in raising the metal to a high temperature and 
 allowing to cool, either rapidly or slowly. At different temperatures 
 metals behave in different ways. 
 
 Every malleable compound of iron, containing the ordinary 
 elements of that metal, which is obtained either by the union of pasty 
 masses of the iron or by any process involving fusion, and which 
 cannot be hardened by an ordinary method, will be called by us 
 " wrought iron." 
 
 Ductility is the property which enables metals to be drawn into 
 wire. It generally decreases with an increase in temperature of 
 the wire at the time of drawing ; but there is no regular ratio between 
 the two. Iron is less ductile at 100° C. and more ductile at 200° C. 
 than it is at 0° C. 
 
 Tenacity is the property possessed by metals, in varying degrees, 
 of resisting the separation of their molecules by the action of a tensile 
 stress. 
 
 Toughness is the property of resisting the separation of the mole- 
 cules after the limit of elasticity has been passed. 
 
 Hardness is the resistance offered by the molecules of a substance 
 to their separation by the penetrating action of another substance. 
 Great differences are observable between the hardness of various 
 metals. 
 
 Elasticity is the power a body possesses of resuming its original 
 form after the removal of an external force which has produced a 
 change in that form. The point at which the elasticity and the 
 applied stress exactly counterbalance each other is termed the 
 " limit of elasticity." If the applied stress were then removed, the 
 material acted upon would resume its original form. If, however, 
 the stress were increased the change in form would become per- 
 manent, and permanent set would be effected. Within the limit 
 of elasticity, a uniform rod of metal lengthens or shortens equally 
 under equal additions of stress. If this were the case beyond that 
 limit it is obvious that this would stretch the bar to twice its original 
 length, or shorten it to zero. This stress, expressed in pounds or 
 tons for a bar of 1-inch square cross-section, is termed the modulus 
 of elasticity. 
 
IRON AND STEEL 113 
 
 The ultimate tensile strength or maximum stress the material 
 can sustain without rupture, the limit of elasticity, and the breaking 
 stress are the points which usually have to be determined, and these 
 alone will be considered here. In testing a piece of metal, the first 
 point to be determined is the limit of elasticity. When a metal, 
 such as a piece of iron or steel, is submitted to stress by pulling its 
 ends in opposite directions, it stretches uniformly throughout its 
 length. There is, however, in such a solid a limit in the applica- 
 tion of the stress up to which the metal, if released, will return to 
 its normal length. This point is the limit of elasticity. 
 
 Influence of Foreign Elements on the Strength of Metals. — The 
 influence of chemical compositions on the mechanical properties 
 of metals is of great importance. The influence of foreign elements 
 is best shown in the case of iron. The properties of this metal are 
 absolutely changed by the presence of a few tenths per cent, of 
 carbon. Phosphorus and silicon produce very dangerous impurities 
 in iron. It is difficult to estimate the influence of silicon. It is 
 known that its addition to molten steel is useful, as it prevents the 
 formation of blowholes in the solidifying mass. 
 
 The colour of metals is influenced by their purity. Thus, iron 
 becomes white by the admixture of carbon, silicon, sulphur, and 
 phosphorus. All metals are fusible. When strongly heated, they 
 pass from a brownish-red to a clear red colour, which gradually 
 increases in luminosity and transparency to a dazzling white. On 
 solidifying from a molten state, metals frequently exhibit efferves- 
 cences due to the expulsion of absorbed gases. This expulsion, before 
 solidification, causes a sudden outburst of metal through the surface. 
 When it passes from the liquid to the solid state, it either does so 
 suddenly, or it passes through an intermediate pasty stage. This 
 fact is occasionally of great metallurgical importance. 
 
 On solidification after melting, metals usually crystallise. 
 The crystallisation of metals is of great importance, as the formation 
 of crystals, due to continued vibration, intense cold, sudden altera- 
 tions of temperature, or the presence of impurities, may render a 
 metal absolutely useless. Welding is the property, possessed by 
 metals which, on cooling from the molten state, pass through a 
 plastic stage before becoming perfectly solid, of being joined together 
 by the cohesion of the molecules introduced by the application of an 
 extraneous force, such as hammering. This property is exhibited 
 in a marked degree by iron and platinum at a white heat. 
 
 All steel is iron, differing only in containing a larger percentage 
 of carbon, usually with a small quantity of silicon and manganese, 
 
 8 
 
114 MODERN METHODS OF WELDING 
 
 and often a small percentage of some other metal. What is known 
 as mild steel is a product that stands between wrought iron and the 
 hardest steel, as that used for making cutting tools. This mild 
 steel has largely taken the place of wrought iron and is used in the 
 construction of steel buildings. Pure iron is a soft, greyish-white 
 metal, very ductile and malleable, and highly tenacious. This is 
 generally used as welding iron. After fusion, pure iron exhibits 
 a crystalline scaly fracture. It is softer than wrought iron, and is 
 not affected by heating to redness and quenching in cold water. 
 It is highly magnetic, and welds readily. Its specific heat is 0T13 
 and its specific gravity 7-675. It melts to a lower temperature than 
 platinum, about 1,600° C. In mass it is unaffected by dry or moist 
 air, and more so in oxygen, yielding a scaly coating of oxide. When 
 molten, it dissolves or occludes various gases in considerable quan- 
 tities. Hydrogen, carbon monoxide, and nitrogen are thus taken 
 up and given out on cooling. 
 
 The above physical properties are present in a greater or less 
 degree in cast or wrought iron and steel, the extent to which they are 
 modified depending on the purity of the substance. These bodies 
 consist of iron containing varying proportions of carbon, silicon, 
 manganese, sulphur, phosphorus, etc. The main difference between 
 the properties of cast and wrought iron and steel are due to the 
 presence of carbon in the metal, depending on the amount and the 
 manner in which it exists in the iron. The maximum amount of 
 carbon taken up by pure iron is stated to be 0-475 per cent. In 
 cast iron containing manganese, a little over 5 per cent, may be 
 present. Steel may contain up to 1-8 per cent., while the carbon in 
 wrought iron seldom exceeds 0-25 per cent, and may fall as low as 
 0-05 per cent. 
 
 The designation of steel was formerly confined to those varieties 
 of iron which could be hardened by heating to redness and plung- 
 ing in cold water. The introduction of the Bessemer process marked 
 a new era. The metal produced by that process lacks the fibrous 
 character associated with wrought iron and partakes more or less 
 of the character of steel. Varieties possessing more than 0-3 per 
 cent, of carbon sensibly harden when treated in the same manner 
 as steel. Some steels are even softer than wrought iron. Since the 
 hardening property is dependent on the amount of carbon contained, 
 a classification based on the percentage of that element is most 
 convenient. Steel containing less than 0-5 per cent, is classed as mild 
 steel. The different nature of the metals may be shown by the use 
 of such titles as Bessemer, Siemens, or open-hearth steel. Some 
 
IRON AND STEEL 115 
 
 of these contain as little as 0-08 per cent, of carbon — less than is often 
 present in wrought iron. (This is the easiest of all steels to weld, 
 but very little of it is manufactured, not being the usual standard.) 
 They differ from wrought iron in being devoid of fibre, more homo- 
 geneous, and, unlike it, are obtained in a state of fusion. 
 
 The fracture of steel becomes finer the larger proportion of 
 carbon present, but it is affected by such treatment as hammering. 
 
 Cold steel of hard temper breaks with a clear, uniform, grey, 
 fine, granular fracture. After hardening, the colour is somewhat 
 whiter. It is very malleable, but requires working more carefully 
 and at a lower temperature than wrought iron. Steel containing 
 less than 1-25 per cent, of carbon can be welded, but at a lower 
 temperature than wrought iron, or the steel will be burnt. To 
 render the surfaces clean at the lower heat, borax mixed with 
 one-tenth of its weight of sal ammoniac should be employed to 
 dissolve the scale. 
 
 The specific gravity of steel varies from 7-624 to 7-813. The 
 melting-point varies with the proportion of carbon. The softest 
 melts a little below 1,600° C. The tenacity varies from 22 tons in 
 mild steel to 70 tons in steel of hard temper. The elasticity exceeds 
 that of wrought iron, while the ductility is equal to the best qualities 
 of that substance. The mild varieties suffer an elongation and 
 diminution in area when subject to a stretching force greater than 
 wrought iron. The elongation of the harder varieties is much less, 
 but the elastic limit is high. Mild steels, when being welded and 
 becoming molten at the weld, are liable to " bod." This is due to the 
 disengagement of dissolved gases, mainly H, N, and CO, which are 
 given up as the metal cools off. The bubbles of the gas cause the 
 metal to be honeycombed and vascular. 
 
 Tenacity of metal is determined by straining a piece of metal 
 of known dimensions, and observing the amount of force necessary 
 to fracture it. Elongation, the extent to which a metal elongates 
 prior to fracture, is a matter of greatest importance. Tough ductile 
 metals show a considerable increase in length ; hard, brittle metals 
 elongate but little. Important evidence as to the working qualities 
 of the material and efficiency of the weld is furnished. To deter- 
 mine the elongation, the test-piece is measured between the points 
 at which it is gripped before and after straining till fractured (it is 
 usual to put two marks on the test-piece for measuring), and the 
 increase is stated in percentage of the original length. Thus a 
 10-inch test-piece of boiler steel measured 12-5 inches after fracture 
 — i.e., 2-5 inches over 10 inches, or 25 per cent. Elongation is 
 
116 MODERN METHODS OF WELDING 
 
 accompanied by a diminution in area of section. This is measured 
 in order to determine whether the elongation was local or uniformly 
 distributed. Sometimes the contraction in area is confined to the 
 region of the fracture. 
 
 Some Difficulties in Welding. 
 
 Welds on mild steel, which apparently are the most easy to 
 obtain, are in reality those which require the greatest study and 
 skill on the part of the operator. The welding must be done to give 
 the weld the mechanical properties approaching those of the metal 
 to be welded. In cases where the process is applied without the 
 knowledge of the technique, the strength of the metal and, above 
 all, its elongation are considerably lowered. In short, the operation 
 of welding can lower, at the line of joining, the principal qualities of 
 mild steel, which are particularly required in metallic construction. 
 
 If the operator will go thoroughly through the chapter on iron and 
 steel, he will learn much that will be of great assistance to him in 
 his welding. One of the important things to remember in welding 
 iron and steel is the formation of the oxide and its inclusion in the 
 metal forming blowholes. It is to be noted that there is always 
 a formation of oxide at the surface of the iron and steel melted under 
 the action of the blowpipe. This oxide fuses before the metal, and 
 is lighter. Therefore it rises to the surface of the metal when molten, 
 and can be eliminated by passing the welding-rod in a horizontal 
 position over the weld while the metal is molten. Steel and iron 
 in a molten state dissolve 1-1 per cent, of oxide. This is always the 
 case, and unless the operator has full knowledge of the technical 
 and metallurgical points, he prevents the joints standing the stresses 
 required. The technique of oxy-acetylene welding is very little 
 known, and there are yet many problems to be discovered. 
 
 Many operators make the frequent mistake of interposing 
 oxide in the weld. In nearly all these cases this is caused by burn- 
 ing of the weld, when an excess of oxygen is used, the flame held 
 on the weld too long, and too large a surface liquid. It causes the 
 weld to become " cinderised." In the melting period of the weld the 
 suspended particles are no doubt due to the spitting of the metal as it 
 fuses, but the origin of the oxide when the molten metal is covered 
 with the slag is less certain. It may possibly be due to the presence 
 of iron vapour in numerous bubbles of carbon monoxide formed 
 throughout the molten weld; or probably to the spurting of the 
 molten metal. The oxide defuses in the molten metal, and reaction 
 takes place with the carbon and manganese and reduces their 
 
IRON AND STEEL 
 
 in 
 
 strength. The dissolving of the gases may be given up, if the 
 operators take care, when the metal is molten, to allow these gases 
 to escape, and to make good solid welds. 
 
 One fault, of which many operators are guilty, is to use a toe 
 high-powered blowpipe, or to use too high pressure of oxygen, whict 
 causes an over-fierce flame. Consequently the metal has not time 
 to get thoroughly hot before it begins to become liquid. Onlj 
 just the surface is "swilled," and the underneath layer is no1 
 anywhere near the welding-point. This is the cause of many defec- 
 tive welds in which there is adhesion and oxide is interposed in the 
 weld. 
 
 a: 
 
 \ r 
 
 Fig. 61. — Tension al Tests, Welded Flat and Bars. 
 
 This can be avoided if the proper-sized blowpipe is used, and the 
 pressure of oxygen is that prescribed by the makers of the blowpipe, 
 and no more. Then the weld must be proceeded with at a moderate 
 speed, giving time for the edges of the article to melt thoroughly 
 into a thick liquid form. Then the feeding-rod must be added and 
 the weld filled up. The blowpipe must not be allowed to remain on 
 the metal after it has been melted (and the tip must be kej3t about 
 yV inch from the metal that is being welded) as it burns the metal 
 after the first melting. If the metal gets too hot, it becomes too 
 liquid and is burnt, and the carbon and other elements are partly 
 destroyed, thereby reducing the strength of the weld considerably. 
 
118 MODERN METHODS OE WELDING 
 
 rhe point of the molten metal should be just a thick liquid, and 
 mould never be made hotter or thinner liquid, which causes oxida- 
 :ion and swilling. 
 
 The author would draw attention again to his previous remarks 
 is to operators testing their work from time to time. Leading 
 mgineers, through so many past failures, and so many bad and 
 defective welds, are not very favourable to allowing oxy-acetylene 
 welding to be done on anything that has to stand any great stresses. 
 rherefore operators should, without fail, make repeated tests from 
 ;ample welded joints, as outlined in previous chapters. In addition 
 :o the tests explained, tensional, distortional, and other mechanical 
 bests can be made, and will greatly assist operators in becoming 
 efficient. These latter tests many schools would be glad to make 
 m their behalf. 
 
 Blowpipes must be the best, if good welding is required. The 
 previous chapter on the subject has laid this down; a further 
 paragraph will help to emphasise their importance. Absolutely 
 bhe best blowpipe must be procured, irrespective of cost. In the 
 selection the following points should be considered: The blowpipe 
 must have a constant clear flame with a distinct white cone, as large 
 is possible and sharp-edged ; must be easy of regulation ; must not 
 back-fire. There must be a proper mixture of the gases at the nozzle 
 Dutlet, and at the correct pressure specified by makers, and no more, 
 rhere must be no excess of acetylene (which carbonises), no excess 
 Df oxygen (which oxidises). The latter in excess is most dangerous. 
 In no case must a hard or steel wire be used for cleaning out the 
 blowpipe nozzle. A piece of copper wire would suit. If the nozzle 
 3f the blowpipe is only minutely enlarged, it causes disarrangement, 
 and the blowpipe does not work satisfactorily. 
 
 Welding-rods for use on iron and steel are composed of pure 
 iron in the form of wire, in the various sizes for the thickness of metal 
 welds. By using a pure iron rod, the line of weld will also be pure 
 (providing the welding has not been oxidised). All metallurgists 
 ire fighting against the inclusion of phosphorus, sulphur, and silicon 
 in the manufacturing of steel and iron ; therefore operators must 
 also fight against allowing these impurities to find their way into a 
 weld. A good flux for iron and steel is as follows : 
 
 \ 
 
 Borax . . . . 3 parts 
 
 Colophony . . . . 2 ,, 
 
 Pulverised glass . . 3 ,, 
 Steel filings . . . . 2 ,, 
 Carbonate of potash 1 part I 
 Hard soap powder . . 1 „ • / 
 
 r Melt in an earthenware vessel. 
 
IRON AND STEEL 119 
 
 One cannot be too particular in preventing these impurities in 
 the welding line. None but the purest rods must be employed. 
 They are to be got if the price is paid. One must not expect to 
 obtain pure charcoal wire without paying for the purity. The extra 
 cost is saved in many ways. The welding is faster, very much 
 stronger, and almost free from oxide. There is no going over the 
 weld twice, and no burning, as it melts freely. Finally, the weld is 
 neater. 
 
CHAPTER XXI 
 CAST IRON 
 
 Cast iron is the cheapest and most abundant form in which metal 
 is met with in commerce. It is fusible at a temperature which can 
 readily be attained ; and, as it receives remarkably clean and exact 
 impressions of a mould, it can be cheaply produced, even in very 
 intricate forms. Its tensile strength, varying from an average of 
 about 7 tons per square inch in common castings to upwards of 
 15 tons with special mixtures, is ample for many purposes. Its 
 crushing strength is greater than any other material, reaching a 
 maximum of about 100 tons per square inch. Being protected by a 
 skin, cast iron resists atmospheric influences better than wrought 
 iron or steel, while, for wearing surfaces for machinery, nothing is 
 superior to cast iron on cast iron as sufficient area is provided. 
 
 Castings are much more easily and cheaply produced than forg- 
 ings, so that the latter are only employed where special requirements 
 or strength and ductility render their adoption necessary. As 
 compared with steel castings, the advantages of cast iron for ordinary 
 uses include not only the cheapness of the original material, but also 
 the diminished cost in the preparation of the moulds, the smaller 
 loss in casting, and the saving of expense and the time required for 
 annealing, which is necessary for steel, but not for cast iron. Iron 
 castings can, therefore, be prepared to meet a pressing emergency, 
 while their fine surfaces, sharp edges, and pleasing appearance 
 recommend them for general use. It must be noted that, in the 
 welding of castings, one must not go over the weld twice, as it has 
 an oxidising effect on remelting, and the portion of the silicon is 
 diminished, while the sulphur is at the same time absorbed from the 
 blowpipe gases. The natural effect of these changes is shown in 
 the condition of the carbon, which, instead of being almost wholly 
 graphitic, is all combined, thus producing a hard, white iron, deficient 
 in tenacity, and brittle. The physical effects produced when cast 
 iron is remelted are thus merely indications of chemical changes 
 which have taken place in the material, whilst the nature of these 
 changes will vary with the composition of the iron employed. 
 
 120 
 
CAST IRON 
 
 121 
 
 Effect of Size and Shape. — -The strength and solidity of a casting 
 are affected by the bulk of metal employed, and by the form of the 
 casting made. When the metal cools in a mould, a crystalline is 
 developed, the crystals forming at right angles to the cooling surface. 
 If this cooling surface be curved, the crystals interlace, so as to yield 
 a strong uniform structure, while, on the other hand, whenever a 
 sharp change of curvature takes place, a plane of weakness is 
 developed. 
 
 Shrinkage of Cast Iron. — -Although cast iron, especially when very 
 grey, expands at the moment of solidification, the subsequent cooling 
 
 Fig. 62. — Tramway Gear Case, Outside Bearing Broken Off, Successfully 
 
 Welded. 
 
 from a red heat to the ordinary temperature leads to a still greater 
 contraction. The shrinkage of castings is, however, by no means 
 a constant quality, but varies with the proportions of the castings 
 and with the character of the metal used. 
 
 Hardness of Iron. — The hardness or softness of cast iron is, in 
 many instances, of the greatest importance, as the metal has to be 
 turned, planed, filed, or otherwise worked with tools. When cast 
 iron has to be turned or otherwise worked, the hardness is of con- 
 siderable importance, while in some cases smoothness of surface 
 and general perfection of the casting are of the utmost moment. 
 Hard cast iron is brittle, deficient alike in crushing, transverse, and 
 tensile strength, and seldom gives good clean castings. 
 
122 
 
 MODERN METHODS OF WELDING 
 
 Fig. 63. — Press Frame Broken on One Side, 
 and Patched with Two 2-Inch Bolts. 
 
 Bolts failed and welding was necessary. 
 
 When cast iron cools from fusion, the carbon may remain 
 uniformly distributed through the mass- — combined carbon — or a 
 portion of it may separate out in scales resembling graphite. The 
 extent to which separation occurs depends on the rate of cooling 
 
 and the quality of the metal. 
 Slow cooling, and thepresence 
 of silicon and aluminium in 
 the metal, favour the separa- 
 tion, while manganese re- 
 tards it. 
 
 When rapidly cooled, 
 nearly all the carbon remains 
 in the combined form. 
 
 Combined carbon hardens 
 the metal, lowers its melting- 
 point, destroys its mallea- 
 bility and welding power, 
 and tends to make it brittle. 
 The extent to which these 
 effects are produced depends 
 on the amount. In white cast iron containing as much as 3 per cent., 
 the metal is brittle, breaks with a silvery-white fracture, melts-more 
 readily, and passes through a pasty stage in fusing. It is extremely 
 
 hard, and this property isper- 
 manent. In steel for cutting 
 instruments, the amount 
 varies from 0-5 to 1-5 per 
 cent. The hardness and 
 fusibility are increased, the 
 malleability and the welding 
 power reduced, in proportion 
 to the amount of carbon 
 present. 
 
 Graphitic carbon is met 
 only in cast iron, and occa- 
 sionally in steel. It reduces 
 the strength of the metal by 
 interposition between the 
 particles, but does not affect 
 the grains of the irons themselves. The generalisation above as to 
 combined and free carbon only expresses part of the truth. When 
 white cast irons, free from manganese, are heated for a long period, 
 
 Fig. 64. — Press Frame Welded at Cost 
 of £9 7s. 6d. Weld 3 Feet Long and 
 1 Inch Thick. 
 
 In use for the past two and a half years. 
 
CAST IRON 123 
 
 at a high temperature, but below fusion, embedded in red hematite, 
 the characteristic brittleness is lost, and they become more or less 
 malleable. 
 
 In connection with the influence of cooling, cast iron obeys the 
 laws which govern other solutions, for it is well known that slow 
 cooling assists the production of crystals, and leads to the production 
 of larger size, while with rapid cooling both solvent and substance 
 dissolved may solidify together. In a similar manner slow cooling 
 tends to produce graphitic carbon, and the slower the cooling the 
 larger the flakes of graphite which separate. Some kinds of white 
 iron may thus be rendered grey by slow cooling, while some kinds 
 of grey iron may be made perfectly white by rapid cooling or 
 " chilling." 
 
 That the carbon which exists in grey iron is in the graphitic form 
 can be proved by many simple tests. Thus, if finely divided white 
 iron be rubbed between the fingers, it is clean to the touch, while grey 
 iron produces a smooth black coaling on the skin, exactly like that 
 due to plumbago. It has been shown that nearly pure graphite 
 can be separated from grey iron. 
 
 All cast irons contain silicon, in quantities varying in ordinary 
 cases from under 0-5 to over 4 per cent. A small addition of silicon 
 eliminates blowholes, and produces sound castings. As soon as the 
 metal is sound, with the least graphite, the greatest crushing strength 
 is obtained. This condition also gives the maximum density. 
 The further addition of silicon leads to graphitic formation, 
 diminishes the brittleness, and gives the greatest transverse and 
 tensile strength. White iron shrinks during solidification more than 
 grey iron, while highly siliceous iron shrinks still more than white. 
 Hence, on adding silicon to white iron, the shrinkage is diminished ; 
 but an excess of silicon, on the other hand, leads to increased 
 shrinkage. Shrinkage appears closely to follow the hardness of cast 
 iron, hard irons almost invariably shrinking most; and as hardness 
 and shrinkage depend upon the proportion of carbon, they may be 
 regulated by a suitable addition of silicon. 
 
 Cast iron is a granular and crystalline compound of iron and 
 carbon, more or less mixed with uncombined carbon in the form 
 of graphite, but never contains more than 5 per cent. It is harder 
 than pure iron, most brittle, and not so tough. The modes of com- 
 bination of the carbon with the metal, as well as the nature and 
 proportion of foreign matters, such as silicon, aluminium, sulphur, 
 phosphorus, and manganese, determine the infinitely varying quali- 
 ties according to the colour, degree of fusibility, hardness, tenacity, 
 
124 MODERN METHODS OE WELDING 
 
 and so on. All cast irons are not available for foundry purposes. 
 In grey cast iron the carbon is mechanically interspersed in small 
 black specks among the lighter particles of metals, the fracture being 
 a dark grey colour, and being of granular or scaly crystalline char- 
 acter. Grey iron is much softer and tougher than white iron, and 
 may be filed and turned. 
 
 In white cast iron the greater part of the carbon is present in 
 the form of a chemical combination — carbide of iron — and white 
 iron is very brittle, and can neither be turned in a lathe nor filed. 
 Grey cast iron requires a higher degree of heat before it commences 
 to fuse, but becomes very liquid at a sufficiently high temperature. 
 It is important to remember this point when castings are being 
 welded, especially when inclined from the horizontal, as the metal 
 would run away from the weld. White cast iron is not so easy tc 
 weld. It does not flow well, is rather pasty in consistence, and scintil- 
 lates as it flows in the molten state. White cast iron is silvery- 
 white, either granular or crystalline, difficult to melt, brittle, and 
 excessively hard. It is a homogeneous chemical compound of iron 
 with from 2 to 4 per cent, of carbon. Granular cast iron can be con- 
 verted into grey cast iron by fusion and slow cooling, whilst grey 
 cast iron can be converted into granular white cast iron by fusior 
 and sudden cooling. Crystalline white iron is harder and more 
 brittle than granular, and is not capable of being converted into 
 grey cast iron. Grey cast iron contains about 1 per cent., or less. 
 of carbon in chemical combination with the iron, and from 1 to 4 
 per cent, of carbon in the state of graphite in mechanical mixture. 
 The larger the proportion of graphite the weaker and more pliable 
 is the iron. 
 
 The following remarks upon some points already described may 
 aid in roughly estimating the quality of cast iron. 
 
 When the colour is uniform dark grey, the iron is tough, providing 
 there is also high metallic lustre. If there be no metallic lustre the 
 iron will be easily crumbled. The weakest sort of cast iron is where 
 the fracture is of dark colour, mottled, and without lustre. The 
 iron may be accounted hard, tenacious, and stiff when the colour 
 of the fracture is lightest grey, with a high metallic lustre. When 
 the colour is light grey, without metallic lustre, the iron is hard and 
 brittle. When the colour is dull white, the iron is still more hard 
 and brittle than in the last case. When the fracture is greyish- 
 white, interspersed with small radiating crystals, the iron is of the 
 extreme degree of hardness and brittleness. 
 
 When cast iron is dissolved in muriate of lime or muriate of mag- 
 
CAST IRON 125 
 
 nesia, the specific gravity is reduced to 2-155. Most of the iron is 
 removed, and the remainder consists of graphite with the impurities 
 of cast iron. The soft grey iron yields to the hie after the outer 
 crust has been removed. The quality of the iron in a melted state 
 is really judged by the practised eye from the nature of the agitated 
 aspect of its surface. The mass of fluid seems to undergo a circula- 
 tion within itself, having the appearance of ever-varying network. 
 
 When this network is minutely subdivided, it indicates soft iron. 
 If, on the other hand, crystals be thrown up in convolutions, the 
 quality of the metal must be hard. 
 
 One of the most important points in the welding of cast iron 
 is not to go over the weld twice, as each time any part is melted 
 more than once, the graphite burns out and leaves white iron, which 
 is hard and brittle. One must remember that every additional 
 melting of cast iron injures, or is likely to injure, its quality as a 
 structural material by the addition of foreign substances. These 
 reduce the value of the coefficient of resistance at rupture and may, 
 or may not, reduce that of ultimate extension. That is, the metal 
 by remelting becomes weaker, and may become more brittle. 
 
 The numerous failures which puzzle operators in the welding of 
 cast-iron articles are often due to lack of knowledge. They have 
 not made themselves acquainted with the metal they are welding. 
 It is a sine qua non that, if they are to become proficient operators, 
 they must have the necessary metallurgical knowledge, so as to 
 know exactly what are the constituents of the metal being welded. 
 The state of the carbon in the metals depends on whether welds are 
 of grey iron. It is necessary for every welder to study thoroughly 
 the causes that facilitate or prevent the precipitation of the carbon 
 in the form of graphite. The rapid cooling of the molten metal in 
 fusion will bring about the combination of the carbon and the iron. 
 This forms white iron, which is hard and brittle, and cannot be 
 machined or filed. 
 
 Welding of cast iron is a simple process, and with an experienced 
 man, who has knowledge — -metallurgical and chemical — of the 
 article he has to weld, failure is impossible. The repair on the fine 
 of welding is generally of a superior quality to the rest of the casting, 
 owing to the metal added being purer and the weld well carried 
 out and free from blowholes. Cast iron consists of metallic iron, 
 together with at least 1-5 per cent, of carbon. It also contains 
 sulphur, silicon, phosphorus, manganese, and other elements in 
 greater or less proportion; but these, as indicated above, may be 
 regarded as impurities. The proportion of elements other than 
 
126 MODERN METHODS OF WELDING 
 
 iron is usually about 7 per cent, of the total weight. Cast iron is 
 fusible at a temperature of about 1,200° C. When cold it is hard 
 and brittle. It is not malleable or ductile, nor can it be hardened 
 or tempered like ordinary carbon steel. Cast iron, when fused, con- 
 sists of saturated, or nearly saturated, solution of carbon in iron. 
 The amount of carbon which molten iron can thus dissolve is about 
 3 J per cent, of its own weight, though the solubility is largely in- 
 fluenced by the presence of other elements. As long as iron con- 
 taining some 3 per cent, of carbon remains in the fused condition, the 
 composition is uniform throughout, and the carbon has no tendency 
 to separate from the metal except with very grey iron. In this case 
 a layer of graphite may be formed. But when molten cast iron is 
 cooled to a temperature at which it begins to solidify, it may either 
 retain the carbon and solidify in a relatively homogeneous form, 
 called white iron, or may, in solidifying, precipitate the greater part 
 of the carbon in the form of small scales of graphite which, being 
 entangled by, and uniformly distributed through, the iron, imparts 
 to it a somewhat spongy nature, and produces the dark colour and 
 soft character met with in grey iron. The condition which the 
 carbon assumes on the solidification of the mass is dependent partly 
 on the rate of cooling, and still more on the nature and quantity of 
 the associated elements. 
 
 In the early stages of the oxy-acetylene process it was generally 
 considered that cast iron and cast steel could not be welded by this 
 process. The author, over fifteen years ago, welded castings by this 
 process nearly daily, and executed some very important castings 
 with success. At this period he found out how valuable and neces- 
 sary was the preheating furnace for counteracting the phenomenon 
 of expansion and contraction. Since then it has been recognised 
 that the only way of getting satisfactory castings welded is by the 
 use of a heating furnace, both for preheating before welding and 
 for annealing after welding, allowing the casting to cool very slowly 
 Failure is almost impossible if this is carried out, provided the weld 
 has been properly executed. 
 
 Of late years operators and employers have been getting more 
 enlightened on welding processes, and most are now providing these 
 necessary appliances. Also all operators are now taking technica 
 courses at the various schools. Cast iron, in the author's opinion 
 is the easiest of all metals to weld. With a little practice, combinec 
 with technical knowledge, operators are soon able to do intricat< 
 jobs; and the welded part, if done well, is generally better than th< 
 other parts of the casting, because the material added by the weld 
 
CAST IRON 127 
 
 ing-rod is purer in its mixture, and should thus be better metal. 
 Cast iron cannot be forged, but articles are cast. They are not malle- 
 able or ductile. The majority of castings in iron should be capable 
 of being worked — that is, they must be soft metal when finished 
 and able to be filed or machined. The metal of grey castings is 
 usually grey iron. The description at the beginning of this article 
 should be studied thoroughly. One must remember that the rapid 
 cooling of the metal in fusion brings about the combination of the 
 carbon and the iron, which means the formation of white iron. 
 Slow cooling and reheating bring about precipitation of the carbon 
 and grey iron. 
 
 The melting-point of cast iron is 1,200° C, but its oxide melts 
 at 1,350° C. This is an important point, and in melting, the oxide 
 usually flows on the top of the weld, where it can be removed either 
 by a flux or (if the welder be experienced) by the welding-rod, 
 scraping it over the surface horizontally. If the flame is held too 
 long on the casting after the metal is molten, the metal burns and 
 oxidises, the silicon is volatilised, the carbon decarbonised, and the 
 weld will be hard and cannot be machined or filed. When cast- 
 iron articles are welded the welding flame causes, to a certain extent, 
 a volatilisation of the silicon, which is contained in the metal in 
 proportion from 0-5 to 4 per cent., generally about 2 per cent. 
 If this is burnt out, it is necessary to replace the loss, which is done 
 by the welding-rod, which should contain 5 per cent, of silicon. 
 Therefore, as welding takes place, the rod with the increased per- 
 centage of silicon gives back to the casting the percentage that has 
 been volatilised out. This material usually leaves the welded line 
 soft, so that it may be machined or filed. 
 
 Of the difficulties experienced, the first is the expansion and con- 
 traction. Owing to the metal being non-ductile, and devoid of those 
 elements for elongation and elasticity, castings are difficult to handle. 
 The only remedy is preheating in a furnace, and annealing after 
 welding. 
 
 The second difficulty is to prevent the line of welding becoming 
 hard. To stop this the weld must be made at the first trial, and 
 must not be gone over a second time. The welding must also be 
 sharp, as, if the blowpipe is kept on too long, it is burnt and the 
 weld hard. 
 
 Thirdly, the temperature of the casting must be brought up in 
 the furnace to between 850° and 900° C. before any welding takes 
 place. If it is lifted from the furnace to the welding table, it 
 must be done quickly, and quickly welded, and returned to the 
 
128 MODERN METHODS OF WELDING 
 
 annealing furnace before the heat of the casting gets below the 
 850° C. Below this temperature the forces of expansion and 
 contraction begin to come into action, and internal strains will 
 at once be set up, which may, in a few minutes, if the tempera- 
 ture is much below 850° C, crack or fracture in the casting; it is 
 not necessary in the weld. In some cases, the internal strains remain 
 in the castings without fracture, until the casting becomes quite 
 cold, when the fracture from the release of the internal strains occurs. 
 
 After welding, and putting the casting into the furnace, it must 
 at once be seen to that the temperature of the furnace is raised (if it 
 is not already raised as it should be) to 950° C. This is so that the 
 casting after welding can be raised to this heat quickly, to bring 
 the parts of it up to one heat, since the part that bad been welded 
 was very much hotter than the part not welded. Thus one can stop 
 uneven contraction when the casting is being cooled off. After 
 the heat has been raised to 950° C, which would not be long if the 
 furnace is working as it ought to, the furnace must be cooled right 
 down to cold before removing the casting. 
 
 Some illustrations of preheating and annealing are given in 
 previous chapters relating to expansion and contraction. They 
 should be carefully studied. 
 
 Fourthly come the difficulties of lack of penetration, bad joining, 
 sinking of the surfaces, blowholes, and interposition of oxide. 
 Lack of penetration is a frequent occurrence. There is, however, 
 no justification for it, if operators will only go to the bottom of the 
 weld in all cases. The blowjnpe must be kept on the particular 
 welding line, until such time as the bottom is melted. When the 
 bottom is found at the starting-point, there should be no mistake 
 about the line being continued from the bottom of the weld. 
 
 Interposition of oxide is a common occurrence, but it is one which 
 can easily be avoided. It occurs through using an excess of oxygen, 
 making the molten metal oxidised too liquid and too hot, and 
 through being too long on welding and going over it more than once. 
 The oxide forming through these errors is imprisoned in the metal. 
 Therefore, it is not homogeneous, and the weld is defective. This 
 will not come about if the operator will first of all bevel the joint 
 where it has to be welded. When it is ready and heated, welding 
 must take place at once by commencing on the bottom of the bevel. 
 As soon as this is melted, add the welding-rod, previously heated, 
 in the bevel, move the blowpipe forward with an elliptical sweep, 
 keeping it close in the line of metal. Do not let the white tip touch 
 the metal, but keep it about ^ inch from it, and go steadily and 
 
CAST IRON 
 
 129 
 
 slowly forward, filling up the bevel uniformly with the feeding- 
 rod until the end of the weld is reached. There must be no stoppage 
 whatever, while welding the line prepared. If you stop, your 
 weld will be hard and cannot be filed or machined. If it is done 
 quickly, and filled in as the welding proceeds, with no stoppage, 
 the weld will be a success, neat, strong, and easy to work. 
 
 If the article to be welded is a complete casting with a break, 
 then it would be necessary to bevel the fracture on both sides before 
 welding: such a case is a motor cylinder broken on the flanges. 
 The one illustration below would require bevelling; and take care 
 that the weld is thoroughly penetrated. 
 
 Welding-rods used for cast iron should be made more or less 
 from good tough metal, 
 with a fine granular 
 structure, and free from 
 impurities, especially 
 manganese. They 
 should be incorporated 
 with silicon, in two 
 grades, one to have 2-9 
 per cent, and the other 
 to have 4-1 per cent, 
 of silicon. The object 
 of this grading is very 
 important, as the one 
 has a large granular 
 structure, and the other 
 
 has a fine granular structure. The latter would be used on heavy 
 machinery work, and the former on smaller castings, such as motor- 
 car cylinders, which are cast from fine granular material. 
 
 All founders grade their metal in this manner. They do not 
 use the same for both light and heavy castings. Therefore it 
 is imperative that the welding-rods should be made in accordance 
 with the general specifications of the general castings; and they 
 should be very smooth and regular in thickness, which should be 
 from | inch diameter rising by T X g inch diameter to | inch diameter. 
 
 They should be 20 inches long, and should be all sand-blasted, 
 after they are cast. This is an important point. Otherwise the rods 
 will be all rough and covered with sand, which is very detrimental 
 to the weld. No welding -rods must be used which have not had 
 the moulding sand removed; nor must welding-rods be used that 
 have been cast in solid moulds, as they would be chilled and hard. 
 
 9 
 
 Fig. 65. — Two-Cylinder Motor Engine Broken. 
 
130 
 
 MODERN METHODS OF WELDING 
 
 It is not agreed in the welding trade that fluxes are necessary 
 for the oxy-acetylene welding of cast iron. The author does not 
 advise the use of any flux whatever, in general. But there are a few 
 cases in which a flux is of great assistance — e.g., where castings are 
 old, or are full of sand and slag. This will only occur where cheap 
 castings are made. The method of using the flux is to dip the 
 end of the rod into flux, which should be near to hand and to the rod 
 being heated. The flux must not be thrown into the molten weld, 
 as too much would make the weld hard, so that it could not be 
 worked. 
 
 The illustration (Fig. 67) shows a good test of cast-iron welding 
 — a water pump accidentally broken while being worked in the 
 
 shop. It was welded in thirty-six 
 minutes. 
 
 The method of execution was as 
 follows: The casting was first bevelled 
 with a diamond-point chisel, the metal 
 being ~- tJ inch thick. The gas furnace 
 had already been lit. The casting was 
 placed on a sliding grid, which fitted on 
 the sides of the furnace. This sliding 
 grid, with casting placed and fixed in 
 position, was lifted by a small swinging 
 hand-crane, hoisted by a rope, and 
 swung round to the furnace, when the 
 grid was pushed into the furnace. The 
 casting was covered with an asbestos 
 "shawl," and the door closed. The tem- 
 perature was then at 550° C. Another 
 burner was lit, and the temperature rose to 950° C. in thirty-five 
 minutes Half the burners were put out, and the sliding grid, with the 
 casting still fixed, was lifted by the crane and placed on the welding 
 table. The blowpipe, already in position, was lit, and the welding 
 started, and was continued without interruption or without going 
 over any part of the weld twice. The grid was then attached to the 
 crane, which swung the casting back to the furnace. The door was 
 closed and the two burners lit till the temperature (then 800° C.) was 
 raised to 950° C. (this occupied only thirty-six minutes from the 
 time the furnace door opened to take it out till the door closed for 
 annealing). Then the burners were turned right out, and the cast- 
 ing was allowed to cool slowly overnight. The result was a first- 
 class job, strong, well-annealed, and with no distortion. 
 
 ^Hb£?B KlM 
 
 Fig. 66. — Two Cylinders 
 Welded Complete. 
 
CAST IRON 131 
 
 The field for welding such castings is enormous. The demand 
 is greater than the supply. There are few operators able to do all 
 jobs as they come along. If they had scientific and technical train- 
 ing, which should be followed up with practical work, they would 
 become proficient, and would be able to do any class of work. The 
 possibilities of welding broken castings are great, and almost any 
 repair can be done. A few articles here are shown which can be and 
 have been welded by this process. Good jobs can always be done if 
 
 Fig. 67. — Broken Two-Cylinder Water Pump. 
 
 operators carry out carefully the instructions given. I may repeat 
 that in all cast-iron weldings, if success is to be attained, the article 
 must first be prepared, must be preheated (free from air) to a tempera- 
 ture of 950° C, welded with the purest welding-rod, returned to the 
 annealing furnace when the casting gets down to 850° C, whether 
 the casting is finished or not, and, if it is necessary to weld further, 
 must be brought up again to a temperature of 950° C. It can be 
 taken out for welding, and afterwards returned to the furnace and 
 allowed to cool slowly, free from air. 
 
CHAPTER XXII 
 DISSOLVED ACETYLENE 
 
 Dissolved acetylene, sometimes called high-pressure acetylene, 
 is being greatly used, owing to its convenience, the purity of its 
 acetylene, the equal pressure and proper mixture of the two gases. 
 the regularity of flame, the absence of oxidising or carbonising, the 
 well-maintained sharp cone at the tip, and the continuous welding 
 till the cylinder is empty. 
 
 Dissolved acetylene is acetylene in its purest form, compressed 
 after purification into cylinders containing an absorbent material. 
 The gas is first generated in an ordinary carbide-to-water type in 
 which the carbide falls in large volumes of water in order to prevent 
 overheating of the gas and to maintain within the generator a tem- 
 perature considerably below that at which local polymerisation of 
 the gas can occur. Otherwise the heat of dissociation, which repre- 
 sents 1«65 per cent, of the total heating value of the gas, and is 
 responsible for the phenomenal heating value of the acetylene flame, 
 is liberated either wholly or in part during generation, and is there- 
 fore not available in the flame ; in that case further local polymerisa- 
 tion takes place, and the tarry residues mix with the water and the 
 gas and make the latter impure. The acetylene, after generation, 
 must be treated by numerous processes in order to extract the impuri- 
 ties. These impurities are in three forms — gaseous, liquid, and solid. 
 There is no single process capable of dealing effectively with the 
 three. The gases must be treated with at least six processes of 
 purification, two of which must be mechanical, four chemical. 
 All chemicals must be incapable of producing overheating of the 
 acetylene undergoing purification. Special precautions must, at 
 all times, be maintained to preclude the possible admixture of air 
 with the acetylene. All purifiers and the acetylene, after purifica- 
 tion, should be tested twice daily in order to secure complete 
 purification. 
 
 The acetylene is then compressed into cylinders already prepared 
 with solvent material. The compressors used for this purpose are 
 multiple stage compressors, with intermediate cooling between 
 
 132 
 
DISSOLVED ACETYLENE 
 
 133 
 
 each stage. The cylinders of the compressors must be water- 
 jacketed, and the degree of compression in each stage must not be 
 capable of raising the temperature in excess of 100° C. — that is, 
 one-sixth of the critical temperature. The gas, after compression, 
 and prior to entering the cylinders, must undergo mechanical separa- 
 tion for the extraction of final traces of moisture. 
 
 Cylinders in which dissolved acetylene is stored are made from 
 the finest quality steel, and is cold-drawn to the shape of the cylinder 
 from a flat piece of steel, whilst the bottom and walls are folded over 
 one another and pressed together under a pressure of several tons 
 per square inch. Only the finest and most 
 ductile quality of steel would permit of 
 cold-drawing for 44 inches in length. The 
 cylinder walls, only f\ inch thick, although 
 only intended for use under a pressure of 
 about 200 pounds per square inch, are made 
 to stand a pressure of 1,000 pounds per 
 square inch. As is well known, acetylene 
 gases can only be stored under pressure 
 with safety when dissolved in a solvent 
 such as acetone, which must be absorbed 
 by a porous material adapted to fill the 
 space in the containing vessel. 
 
 Government regulations require that 
 the porosity of the absorbent material in Fig. 68 
 the container shall not exceed 80 per cent., 
 and shall be homogeneous through the 
 material without any free gas space. In 
 the past it has been customary to employ, 
 as the absorbent body granulated, solid 
 material such as charcoal, or an animal 
 filament, such as silk, but these substances 
 are subject to certain disadvantages. Granulated charcoal and other 
 solid materials tend to disintegrate into dust by attrition of the par- 
 ticles, if the container is subjected to repeated vibration or bumping. 
 This disintegration creates free gas space and also dust particles, which 
 are liable to blow forward with the gas and block the container valve 
 or the nozzle of the blowpipe employed when the dissolved acetylene 
 is used for welding. Silk, owing to its fibrous nature, does not dis- 
 integrate, but, on the other hand, it is not sufficiently resilient to 
 obviate packing, and saturated with the liquid solvent and subjected 
 to repeated vibration and bumping. The consequence is that free 
 
 Perfect Neutral 
 Flame. 
 When blowpipe is working 
 properly, the length of the 
 small white cone is as shown. 
 In the patent " D.A." blow- 
 pipe, the numbers on the tip 
 correspond to the consump- 
 tion of acetylene in litres per 
 hour. 
 
134 MODERN METHODS OF WELDING 
 
 gas space may be created after the container has once been packed 
 to the prescribed porosity and put in use. 
 
 There has since been patented an improved method of storing 
 compressed acetylene gas. This patent was taken out by Thomas 
 Gaskel Allen, of London, July 27, 1916. The object of the invention 
 is to provide a new type of filling material, superior to that used in 
 the past. According to the material employed, it is sometimes 
 known as "kapok " (Javanese fibre), or Indian kapok. One form 
 suitable for the purpose of this invention is the Eriodendron 
 anfractuosum, but the invention covers any suitable variety of 
 kapok. Using this material, a much smaller weight than previously 
 is necessary to obtain a porosity of 805 in the container. 
 
 Further, kapok has a tendency to swell when it absorbs the liquid 
 solvent, thus precluding altogether the possibility of free gas space 
 forming within the container after it has been once packed to the 
 required porosity. On examination with the microscope, the central 
 tube, filled with air, gives to fibres or kapok its very valuable light- 
 ness. The fibres are absolutely impermeable by water, owing to 
 the presence of wax with which the filaments are coated. They will 
 support from thirty to thirty-five times their own weight in water. 
 Ordinary cork will only float five times its weight. This special fibre, 
 when used as a filling material, cannot disintegrate into dust, no 
 free gas space can be created by constant vibration, and no dust can 
 pass through the blowpipe. 
 
 Dissolved acetylene compressed in cylinders provides a definite 
 volume of acetylene of the highest purity, at a constant pressure, 
 controlled by a regulator fixed on the oxygen cylinder valve. This, 
 together with the oxygen in equal volumes and at equal pressures, 
 gives a wide range for welding. The clear flame obtained is shown 
 in Fig. 68. 
 
 The cone should not be allowed to touch the metal, but should be 
 held so that the required heat is obtained without burning the work. 
 When stopping the blowpipe, always turn off the oxygen first. 
 This system has many advantages over the low-pressure process. 
 The gas is always ready for use without waste of time in preparation, 
 and when shut off it can be stored indefinitely without loss. It is 
 handled with the greatest ease, and can be used in any position. 
 The apparatus is cheap ; the handling of carbide and water and the 
 necessity of removal of residue are eliminated. There is no danger 
 to the operator, and no bad smell, owing to the purity of the gas. 
 
CHAPTER XXIII 
 CUTTING IRON AND STEEL 
 
 The use of oxygen for cutting iron and steel is being developed 
 enormously, and is being adopted everywhere. The method con- 
 sists essentially in an ordinary blowpipe, with an additional passage 
 through which an independent and separately controlled stream of 
 oxygen is supplied at the discretion of the operator. This separate 
 supply of oxygen may be discharged through the centre of the blow- 
 pipe, in which case the mixed gases employed for heating are con- 
 ducted through an annulus surrounding it; or the supply may be 
 brought in a passage immediately behind the heating flame. 
 
 The simple expedient of maintaining an independent heating 
 jet in operation, whilst the cutter is travelling, renders the cutting 
 operation continuous. It furnishes the quantity of additional heat 
 necessary to render the oxide fluid, so that it can be blown away 
 through the cut by the separate jet of oxygen. The cutting opera- 
 tion can be mastered by any intelligent workman in a few hours. 
 The edge or surface of the plate at the point to be cut is heated by the 
 mixed jet of oxygen and acetylene. When this spot has been brought 
 to a state lower than the melting-point, a fine jet of oxygen is dis- 
 charged upon it. This immediately produces combustion of the 
 metal, with the resulting formation of the oxide. The jet of oxygen 
 is made sufficiently strong to blow away this oxide in front of it, 
 with the result that a clean, narrow cut is effected through the metal 
 at a speed of travel which is comparable with hot sawing. The 
 metal on each side of the cut is neither melted nor injured in any 
 way, as the action proceeds too rapidly for the heat to spread. 
 In fact, the edges present the sharp and purely metallic surface of a 
 saw cut. 
 
 The cutting may be made to follow any desired lines, executing 
 circles, curves, or profiles as desired, for which purposes guides and 
 other mechanical contrivances are supplied. Special appliances are 
 supplied for ensuring a steady movement of the cutting nozzle, 
 a matter of considerable importance where neat and accurate work 
 is desired. The process may be employed for cutting sections of any 
 
 135 
 
136 MODERN METHODS OF WELDING 
 
 thickness up to 16 or 20 inches. In metal cutting by oxygen, the 
 melting-point of the metal should never be reached. The successful 
 employment of oxygen for this purpose, therefore, depends on the 
 melting-point of oxide being lower than that of the metal. 
 
 The cutters are made in several varieties. The British Oxygen 
 Company are makers of the one illustrated below, which is a 
 " universal "; it is a reliable article and is largely used. 
 
 Rubber tubing must be fixed at and H respectively. The 
 gases for heating are separately adjusted by valves R and H. 
 These mixed gases are discharged through the passages D. The 
 valve regulating the supply of oxygen for cutting is separately 
 controlled by means of the thumb screw Q or the thumb lever valve 
 P. The separate jet of oxygen for cutting is discharged through 
 the passage C. A is an adjustable sliding guide, which can be 
 
 69. — Hand-Cutting Blowpipe. 
 
 attached to the cutter head at B, in order to maintain a uniform 
 distance between the cutter and the work. There are two nozzles — 
 one for cutting up to 6 inches thick, and a spare one to cut over 
 6 inches. The table on p. 137 gives the approximate results with 
 various thicknesses, the consumption of oxygen, and lineal feet cut 
 per hour, etc. 
 
 This is a very useful table. Operators can compare results from 
 their own cutting by it. 
 
 Since the period in which oxygen was introduced the develop- 
 ment has been rapid. Here, again, the war has brought this unique 
 process into vast uses, which would have taken ten years of industrial 
 work to attain. 
 
 The real theory of oxygen cutting of iron and steel was a very 
 simple process, and was known for years before it became a com- 
 mercial proposition. It was found in the chemical laboratory that 
 jf a thin strip of iron or steel is plunged into a jar of oxygen, after 
 
CUTTING IRON AND STEEL 
 
 137 
 
 being heated to redness, the iron ignites to incandescence and 
 burns right away. This, then, is the equivalent to cutting steel by 
 oxygen, as we shall see as we go along. 
 
 Thickness 
 of Plate. 
 
 Size of 
 Gutting 
 
 Oxygen 
 Pressure. 
 
 Consumption 
 
 of Oxygen 
 
 Feet of 
 Metal Cut 
 
 Consumption 
 of Oxygen 
 
 Nozzle. 
 
 per Hour. 
 
 per Hour. 
 
 per Foot Cut. 
 
 Inches. 
 
 Inches. 
 
 
 
 
 
 i 
 
 i 
 
 "3 5" 
 
 24 
 
 48 
 
 65 
 
 0-75 
 
 i 
 
 1 
 7)^" 
 
 28 
 
 60 
 
 60 
 
 1 
 
 3 
 
 J 
 
 . 32 
 
 75 
 
 50 
 
 1-5 
 
 1 
 
 t 
 
 TB" 
 
 32 
 
 88 
 
 40 
 
 3-2 
 
 H 
 
 1 
 
 36 
 
 95 
 
 35 
 
 3-7 
 
 U 
 
 1 
 
 39 
 
 105 
 
 30 
 
 3-5 
 
 2" 
 
 1 
 
 45 
 
 125 
 
 25 
 
 5 
 
 3 
 
 1 
 
 52 
 
 180 
 
 20 
 
 9 
 
 4 
 
 1 
 
 1 G" 
 
 58 
 
 300 
 
 20 
 
 15 
 
 5 
 
 1 
 
 115" 
 
 65 
 
 420 
 
 20 
 
 21 
 
 6 
 
 1 
 Tb" 
 
 70 
 
 432 
 
 18 
 
 24 
 
 8 
 
 "3"5" 
 
 80 
 
 504 
 
 18 
 
 28 
 
 9 
 
 S 
 "5% 
 
 95 
 
 510 
 
 17 
 
 30 
 
 11 
 
 r 
 
 "CT 
 
 100 
 
 620 
 
 13 
 
 48 
 
 12 
 
 7 
 B"T 
 
 125 
 
 650 
 
 13 
 
 50 
 
 Most metals oxidise under the action of the oxygen in the atmo- 
 sphere, as is instanced by the rusting of iron exposed to the air. 
 This is a form of oxide of iron. Upon the heating of the iron, the 
 oxidation is very much more rapid and intense. Considering 
 the method of cutting by a jet of oxygen, the object is to make the 
 oxidation as intense as possible, so as to burn in the shortest time 
 when in connection with the oxygen, and to obtain the narrowest 
 cut possible. The oxidation takes place at the part which has pre- 
 viously been heated to redness, because at this temperature reaction 
 takes place readily. The combustion of this part of the iron dis- 
 engages heat, a portion of which is absorbed by the neighbouring 
 part. This is sufficient to raise it to a red heat, so that it in turn 
 burns, and this reaction is progressively propagated throughout 
 the metal. The oxide formed has a much lower melting-point than 
 that of the metal, and is detached, leaving the iron continually bare. 
 
 Highly carbonised steels, whose melting-points are appreciably 
 lower than that of iron and in the neighbourhood of the oxide of the 
 metal, do not lend themselves to cutting because the oxidation does 
 not propagate itself; nor does cast iron, on account of the impossi- 
 bility of eliminating the oxide mixed with the molten metal. Iron 
 and steel are about the only metals that can be cut by the pxy- 
 
 / 
 
138 MODERN METHODS OF WELDING 
 
 acetylene process. This is owing to the oxide of iron and steel 
 melting at a much lower temperature than the metal itself. Other 
 metals, where the melting-point of the oxides and the metals are 
 nearly equal to one another, cannot be cut by this process. The 
 reason why iron and steel can be cut is because, when the metal is 
 heated to redness, and oxygen impinges upon it, it is immediately 
 oxidised, and the pressure of the oxygen blows the oxide away as it is 
 formed, leaving a clean cut through the metal. Most of the non- 
 ferrous metals cannot be cut by oxygen for the reason stated above, 
 and also very high carbon steels do not lend themselves easily to 
 cutting. They can be cut, but it is very difficult, and many failures 
 occur. Nor does cast iron lend itself to cutting with an oxy-acetylene 
 cutter. 
 
 Cutting by a jet of oxygen can only be applied to iron and steel 
 in a continuous manner by contact with oxygen, for the reason 
 that iron and steel oxidise very rapidly when heated to redness, and 
 the oxide of iron formed by the impinging of the jet of oxygen com- 
 bines and forms in a molten mass on the steel or iron plate; and, at 
 the same time that the molten oxide is formed, the pressure of the 
 oxygen blows the molten mass away, leaving a clear channel through 
 the whole thickness of the plate. In order to make the cut continu- 
 ous, it is necessary to maintain the steel plate at a red heat along the 
 line of cutting. Taking into consideration that the plate absorbs 
 a quantity of the heat by conducting it over the cold parts of the 
 plate, the cutting pipe has to have a preheating flame, in addition to 
 the jet of oxygen for blowing away the oxide as it is formed. 
 The cutting pipe must be regulated for speed by the heating 
 of the plate, and this heating must be continued as the cutting pro- 
 ceeds. It is not necessary to heat the plates to more than red heat 
 if a clean cut is desired. By raising the temperature to a welding 
 heat, cutting action is stopped, and the metal becomes blobbed and 
 forms a clot of oxide on the top of the plate. 
 
 One must try to reduce the oxygen to a minimum by regulating 
 the supply. It is not necessary to have high pressure or a large 
 nozzle, except when cutting very thick stuff, because excess of 
 oxygen means that the oxide will contain an excess of oxygen and 
 the resulting oxide will be Fe 3 4 — i.e., three of iron and four of 
 oxygen. If the oxygen is correctly regulated and the pressure not 
 too heavy, the cutting will be much more economical, and more 
 lineal feet will result, with cleaner cuts. The oxide formed would 
 be FeO — i.e., one of iron and one of oxygen. The pressure of oxygen 
 required to cut the various thicknesses must be carefully considered. 
 
 
CUTTING IRON AND STEEL 
 
 139 
 
 In many cases it is very indeterminate. What is desirable is the 
 thinnest possible jet, with great length, capable of blowing the oxide 
 away as it is formed, leaving the narrowest clean cut, and using 
 oxygen at the very lowest pressure. The heating flame — its intensity 
 and length, the distribution of heat— always influences the result, 
 and its direction and distance are most important. 
 
 Also, the purity of the oxygen is most important. At 99 per 
 cent, purity, the maximum lineal feet of cutting can be obtained; 
 95 to 96 per cent, of purity reduces the lineal cutting from 10 to 
 15 per cent. With 90 per cent, cutting will only be intermittent. 
 
 Fig. 70. — The Radiograph Cutting Steel Plate with the Oxy-Acetylene 
 Flame, at Speeds Varying from 18 Inches to 2 Inches per Minute on 
 Plate from i Inch to 20 Inches Thick. 
 
 It will be noted that practically any thickness can be cut, from the 
 thinnest to 16 or even 20 inches thick. With thick plates machines 
 are needed to guide the line of cutting. Fig. 70 illustrates cutting 
 a large plate. 
 
 Where cutting has to be done by hand, care must be taken that 
 the line of cutting is straight. In many cases, rollers are used to 
 steady the cutter and to act as a guiding support. Operators 
 must see that the heating flame and the delivery of the oxygen for 
 cutting are regulated proportionally to the thickness of the plate to 
 be cut. The pressure should be regulated at the reducing valve, 
 and the minimum pressure must be used in all cutting operations, 
 as a cleaner cut will be got. If a high pressure of oxygen is used 
 
140 
 
 MODERN METHODS OF WELDING 
 
 when cutting iron or steel, it spreads around the place where heating 
 is taking place; hence it cools the surface, thereby retarding the 
 cutting. 
 
 It is not necessary to have increased pressure of oxygen. Many 
 welders think that the higher the oxygen pressure, the quicker 
 and better. This is a fallacy. Pressures from 10 to 40 pounds are 
 
 Fig. 71. — Coupled Cylinders, for Continuous Work. 
 
 ample for most commercial thicknesses, and more cutting and neater 
 work will be obtained if the oxygen is kept at these pressures. 
 Further, the saving in oxygen is large — quite 20 per cent. It is' not 
 pressure that is needed, but volume. When cutting a great thick- 
 ness, several cylinders are brought into requisition and coupled up 
 together. From this one gets a large volume; but it only needs a 
 medium pressure. The above illustration shows these coupled by 
 cylinders. 
 
CUTTING IRON AND STEEL 
 
 141 
 
 This arrangement enables three cylinders at a time to be de- 
 tached when empty, and replaced with full ones, while any or all 
 of the others are feeding oxygen to the regulator fixed on the tripod 
 stand in the centre. 
 
 00 £ 
 
 O £ 
 
 & < 
 
 Hi # 
 
 Q S 
 
 ^ s 
 
 Manufacturers should use the oxy-acetylene cutting process 
 much more than they do at the present time. They will find impor- 
 tant advantages through operating the torches mechanically, when 
 such a procedure is practicable. More notable among the benefits 
 secured from mechanical control are increased production from the 
 
142 
 
 MODERN METHODS OF WELDING 
 
 cutters and greater uniformity of work. Experience is required to 
 enable the hand operator of a cutter to produce uniform cuts. In 
 shops, the use of mechanically operated cutters eliminates the 
 personal equation and generally improves the quality of the 
 work. 
 
 Figs. 72, 73, and 74 are remarkable and wonderful machines. 
 You will note the ease with which the cutter is handled, also 
 the clea,n-cut edge which it leaves, thereby saving machining. 
 
 Fig. 73. — Circular Cuts in Steel Plate 2\ Inches Thick, with the Radio- 
 graph, at a Speed of 6 Lineal Inches per Minute. 
 
 Inside and outside diameters of these flue sheets for special heaters were cut with 
 the Radiograph and the oxy-acetylene torch. 
 
 It is also the means of increasing production, as cutters are moved 
 at the correct rate, which is predetermined by the feed mechanism. 
 
 Various mechanically operated equipments are designed for 
 particular requirements. There are also standard designs of mechan- 
 ical cutters and welding blowpipes, which are being used with great 
 success. Provision can be made for the mechanical control of cutters 
 adapted for use on parts of a variety of different shapes. 
 
 The three illustrations show that, in addition to making straight 
 cuts on flat plates, it is quite feasible for cutters to follow irregular 
 
CUTTING IRON AND STEEL 143 
 
 outlines on flat plates, or to make cuts on cylindrically shaped 
 pieces. 
 
 With many of these mechanical equipments, the cutters are 
 guided in such a way that they follow the line upon which it is 
 desired to make a cut without calling for special attention from the 
 operator. As previously mentioned, the provision of power drive 
 sets the pace of the cutter, so that it is fed to the work at a prede- 
 termined speed suitable for the thickness of the metal that is being 
 cut. Figs. 72 and 73 show two mechanically worked cutters for 
 cutting ragged edges for such pieces as boiler heads, which have 
 a flange drawn up around the circumference. In Eig. 72 the 
 cutter is shown trimming the flange off a boiler front which is made 
 of steel plate 1| inches thick. It will be apparent from the illus- 
 tration that the cutter is supported by a head carried on a radial 
 arm, on a machine which is so designed that the cutter head can be 
 traversed in either direction on the radial arm, and this arm can be 
 swung round the column of the machine. This combination of 
 movement enables the cutter to be fed round the edge of cylindri- 
 cally shaped pieces, as shown. The cutter head is furnished with 
 the wheels, which engage both inside and outside of the work to 
 provide for holding the point of the cutter in proper relation to the 
 plate it is cutting. The cutter head is worked by the electric motor 
 seen above the radial arm, which transmits power to the wheels on 
 the head, so that they serve the double purpose of holding the cutter 
 in the proper position, and feeding it over the work at the proper 
 rate. The plate being 1J inches thick, it cuts at a rate of 1 foot per 
 minute, and leaves a sufficiently good finish, so that no subsequent 
 machining operation is required. The Radiograph is a portable, 
 motor-driven machine, combining carriage and driving mechanism 
 with motor attached, oxy-acetylene or oxy-hydrogen cutting torch, 
 and means for guiding the heating or cutting gases along straight 
 or curved lines, at constant speed adjusted according to the thick 
 ness of the plate and size of the tip used. 
 
 In making motors for steam turbines, the New York Shipbuilding 
 Corporation pour the molten steel into the mould, which is made so 
 that the rotor casting stands on end, makes the casting longer, 
 which acts as a riser, that applies pressure and assists in production 
 of a solid casting. Obviously, it is necessary to cut off this surplus : 
 and during the process of machining this steel casting is set up in 
 a lathe. Supported in this way, it is an easy matter to rotate the 
 casting at a suitable speed, so that a cutting blowpipe supported in 
 the proper relation to the work will have the steel casting fed to 
 the cutter, at the proper rate for making this cut. 
 
144 
 
 MODERN METHODS OF WELDING 
 
 The illustration below is the casting described herewith. You see 
 that it is in the lathe, and cut through ; the blowpipe is fixed over 
 the casting. 
 
 This rotor casting is 9 inches thick by 16 feet 6 inches in circum- 
 ference, and the cut was completed by the oxy-acetylene cutter in 
 the remarkably short time of thirty -five 'and a half minutes. A 
 good idea of the quality of the finished surface left by the cutter 
 will be gathered from the illustration. 
 
 Fig. 74. — Body of Steam Turbine Rotok oe Cast Steel. 
 
 The end is being cut off by rotating it in the lathe to feed it to the flame of the 
 cutting torch. It is 9 inches thick. 
 
 Users of cutting and welding blowpipes should make themselves 
 familiar with the properties of both hydrogen and acetylene; but 
 there are some who have not had occasion to investigate the proper- 
 ties of carbo-hydrogen. This gas contains from 85 to 88 per cent, 
 hydrogen, and 15 to 12 per cent, of light hydrocarbons of the higher 
 heating series. It is claimed that less oxygen is required for the 
 combustion of this gas than for either acetylene or hydrogen; and 
 the temperature of the carbo-hydrogen flame is approximately 
 4,800° F. 
 
 In referring to the properties of combustible gases which make 
 
CUTTING IRON AND STEEL 145 
 
 them suitable for use in cutting and welding, confusion frequently 
 arises through a misconception of the relation which exists between 
 the number of heat units per cubic foot of gas and the temperature 
 which can be developed by the combustion of that gas. For the 
 performance of the cutting and welding operations, the number 
 of heat units per cubic foot of gas is a matter of minor importance 
 in determining ability to give efficient service. It is the rapidity 
 with which this heat is liberated to develop a high temperature which 
 determines the suitability of the gas for use in the blowpipe. 
 
 The fact is well brought out by comparing the heat value of carbo- 
 hydrogen with that of ordinary illuminating gas, the former having 
 approximately 480 British thermal units of heat per cubic foot, 
 while the latter has approximately 600 per foot. Despite this 
 fact, illuminating gas is unsuitable for use in a blowpipe, because, 
 although it has 25 per cent, more available heat per unit volume of 
 gas, this heat is not liberated rapidly enough to generate a tempera- 
 ture suitable for the cutting of metals. 
 
 It is claimed by the Carbo-Hydrogen Company that this gas has 
 readily cut armour plate up to 24 inches thickness, using standard 
 apparatus, and that open-hearth steel pit castings 36 inches thick 
 across have been cut in half by a special apparatus, using carbo- 
 hydrogen as the combustible gas. For this work a number of 
 oxygen cylinders were manifolded together and a preheating flame 
 was used, which was provided by a blowpipe fitted with a J-inch 
 gas pipe, taking oxygen from the manifold supply direct. Of course, 
 this was an unusual operation, and the cut was not made by one 
 continuous traverse of the cutter. In the case of the 24-inch armour 
 plate this was made in one operation. The composition of the carbo- 
 hydrogen gas is such that an accurate, clean cut is made, and the slag 
 produced is almost pure oxide of iron, there being very little pure 
 iron in it. This indicates that the cutting is accomplished by a 
 complete process of oxidation, which is the ideal method, and not 
 by melting the iron. Where metal is severed by melting it is almost 
 inevitable that a ragged surface should be left in making a cut, so 
 that it is necessary to employ some subsequent process of finishing 
 before the work is ready for use. 
 
 In this chapter information has been presented concerning miscel- 
 laneous applications which have been made of mechanically operated 
 cutting and welding blowpipes, and attention has been called to 
 the benefits secured through the substitution of mechanical con- 
 trol of hand operations. Despite the increased production and 
 higher quality of the workmanship secured through mechanical 
 
 10 
 
146 
 
 MODERN METHODS OF WELDING 
 
 operations, many manufacturers are continuing to use cutters moved 
 over the work by hand. There is no denying that the latter method 
 is capable of producing highly satisfactory results when the cutters 
 are placed in the hands of skilled operators ; but the average mechanic 
 will frequently fail to produce good work until he has had some con- 
 siderable amount of experience. 
 
 It is well worth while for the manufacturer who has use for the 
 
 Fig. 75. — Felling a Stack with an Oxygen Cutter. 
 
 oxy-acetylene torch to investigate carefully the requirements of his 
 work, with the idea of determining whether it could not be handled 
 by one of the standard cutting or welding machines. 
 
 If investigation shows that the work could not be handled on any 
 of the available commercial equipments, the next step is to ascertain 
 whether a special machine using standard cutters could not be deve- 
 loped at a reasonable expense. If so, development of such an equip- 
 ment will usually prove a highly profitable investment, both from 
 
CUTTING IRON AND STEEL 
 
 147 
 
 the standpoint of direct earnings and also by making an improve- 
 ment in the quality of workmanship where cutting or welding opera- 
 tions have to be performed. 
 
 Fig. 75 shows an unique application of the cutting blowpipe, the 
 stack cut through at the bottom at an angle to set the fall. 
 
 Remarkable results, which have revolutionised many methods of 
 
148 MODERN METHODS OF WELDING 
 
 steel cutting, are being obtained with motor-driven machines. 
 One such machine is sold under the name of " oxygraph." By 
 following a drawing with a motor-driven tracer, steel to the thickness 
 of several inches is cut out accurately .in intricate forms. The 
 motor is very small and compact and requires very little power. 
 It can be driven either by battery or by wire attached to an electric- 
 light fixture. Being motor-driven it moves with a uniform speed, 
 and corners and curves are cut with great exactness. 
 
 It will cut steel plate from 1 to 15 inches or more in thickness, 
 it cuts with a narrow smooth kerf, along straight lines, sharp 
 angles, or curves, according to the drawing or pattern. The panta- 
 
 Fig. 77- — Tracer Wheel, Swivel Standard, Rheostat and Electric Motor 
 with Speed Controller of No. 1a Oxygraph. 
 
 graph principle is employed with a motor-propelled tracing wheel, 
 with which the lines of the drawing are followed and reproduced with 
 the cutting torch. The only power required is for revolving the 
 tracing wheel, and this is supplied by a small motor attached to the 
 tracing head, which may be connected to the ordinary electric light 
 or power circuit. The speed of the cutting varies from 2 to 18 inches 
 per minute. 
 
 Machine torches of great power have been developed for oxy- 
 acetylene and oxy-hydrogen cutting with the oxygraph that operate 
 successfully on the heaviest work. The adjustment of the cutting 
 flame is easily learned and skill in the operation of the machine 
 
CUTTING IRON AND STEEL 
 
 149 
 
 is soon acquired by inexperienced operators. An operator capable 
 of running a drilling machine should be able to work the oxygraph 
 efficiently after a few days' instruction. 
 
 The oxygraph has wide application and many uses in tool shops, 
 manufacturing plants, locomotive works, shipyards, drop-forging 
 concerns, and wherever tools, dies, forgings, and shapes are produced. 
 The cutting action of the torch flame is smooth and rapid, and as any 
 
 Fig. 78. 
 
 -Machine Cutting Torch with Motor Control Switch, and Rack- 
 and-Pinion Vertical Adjustment. 
 
 shape can be cut it is comparable to a metal bandsaw of great power, 
 capable of cutting steel 15 inches thick at the rate of 4 or 5 inches 
 per minute, following straight lines, curves, and angles, acute or 
 obtuse. Thin sections are cut more rapidly, of course. 
 
 Punches, dies, stripper plates, and bolsters are cut in tool shops 
 with the oxygraph with resultant saving of time and cost, reaching 
 to 500 per cent., and even more, saving in some cases 
 
150 
 
 MODERN METHODS OF WELDING 
 
 The usual practice in making a cutting or trimming die is to plane 
 the block on the top and bottom, bevel the sides, lay out and drill 
 holes to the line, cut out the walls between the drilled holes with a 
 
 Fig. 79. 
 
 -Small Solid End Connecting-Rod and Billet from which it was 
 Cut on the No. 1a Oxygraph. 
 
 broach or drift, and finish with a hammer, chisel, and file, or by back- 
 ing out on a shaper or slotter. Almost invariably a die made in this 
 manner warps and twists in the process, and requires either re- 
 
 Fig. 80.- 
 
 -Leather-Ctjtting Punch for Shoe Manufacture and Tool Steel 
 Piece from which it was Cut. 
 
 planing on the bottom or shimming up in the bolster. The removal 
 of the mass of steel in the centre of the die relieves internal stress 
 and lets the block warp out of shape. Not so when the rough die 
 block is cut out with the oxygraph. All troubles of this sort are 
 
CUTTING IRON AND STEEL 
 
 151 
 
 eliminated and the danger of cracking in hardening is reduced to a 
 minimum. The rough die block is preheated and cut before planing, 
 using a paper drawing to guide the tracer wheel. When the open- 
 ing has been cut, the die, still hot, is placed in an annealing box, 
 covered, and left to cool. When cold it is planed and finished in the 
 usual manner with the assurance that internal strains have been 
 relieved. Finishing the die to precise dimensions and backing out 
 for clearance is done in the usual manner. 
 
 Not only is the oxygraph useful for blocking out dies but it may 
 be used also to cut the stripper plates and bolsters. The use of the 
 No. 1a oxygraph in a tool shop outlined in the foregoing is one of 
 the many that can be made in the manufacturing plant. It may be 
 
 Fig. 81. — Drop Forged Wrench Trimming Die Roughed out on No. 1a Oxy- 
 graph and Ready to be " Backed Out." 
 
 used for cutting metal templates, cams, patterns, risers for fire 
 escapes, and all shape cutting of any description which comes within 
 the range and capacity of the machine. 
 
 The No. 2 oxygraph is designed for such work as cutting the side 
 frames of mine locomotives, tadpole ends of rudder wings, crank- 
 cheeks of marine engines, mast bands for ships, connecting-rods for 
 locomotive and marine engines, locomotive valve motion links, 
 eccentric rods, and hundreds of parts whose production by other 
 means is slow and costly. 
 
 Figs. 79, 80, and 81 show a few samples of what is possible 
 with these wonderful small machines; those shown are by the 
 oxygraph. 
 
 The above and other illustrations following will serve to give 
 
152 
 
 MODERN METHODS OF WELDING 
 
 some range of work and the speed of performance. But no matter 
 what the shape is, provided the metal is steel or wrought iron 
 
 ^Vw^^ 
 
 No. 1 
 
 No. 1 
 
 No. 1 
 
 No. 3 
 
 w 
 
 No. 3 
 
 Fig. 82. — Views of Three Sets op Dies Cut from 110-Point Carbon Steel, 
 2J Inches Thick, on No. 1 Oxygraph. 
 
 which may be forged to shape, it doubtless can be advantageously 
 cut in the process of fabrication to reduce forging and machining 
 cost. Forge shops use the oxygraph to increase the productive 
 
CUTTING IRON AND STEEL 
 
 153 
 
 capacity of their forging hammers and presses. For instance, a 
 billet weighing many tons may be forged to a roughly rectangular 
 shape, and from that be cut in two, three, or four crank- cheeks 
 weighing, perhaps, 2 tons each. The resultant scrap, in some cases, 
 is of a shape that can be utilised for smaller work by being reforged, 
 hence saving not only time in forging and machining, but metal as 
 well. Cutting may be started at the edge or within the edge of a 
 piece, if conditions require it. The oxygraph torch flame quickly 
 perforates, and thus the cost of drilling and handling is saved. The 
 edges of the cut pieces are square and smooth and, in many cases, 
 no machining is required for finish. If extreme accuracy is required, 
 the cutting can be done so close to the line that machining is a light 
 finishing operation only. These cutting machines are manufactured 
 in America by Davis-Bournonville Company. 
 The following are the costs : 
 
 No. 
 
 Time. 
 Minutes. 
 
 Oxygen. 
 Cubic Feet. 
 
 Acetylene. 
 Cubic Feet. 
 
 Length 
 of Cut. 
 Inches. 
 
 Gas Cost 
 
 at Id. Cubic 
 
 Foot. 
 
 Pence. 
 
 1 
 
 2 
 
 3 
 
 3-5 
 
 4-0 
 4-0 
 
 11-5 
 
 10-5 
 13-2 
 13-2 
 
 36-9 
 
 1-2 
 
 1-4 
 1-4 
 
 4-0 
 
 30 
 34 
 34 
 
 98 
 
 15 
 17 
 
 17 
 
 49 
 
CHAPTER XXIV 
 THERMIT WELDING 
 
 Thermit Process. — This process of welding metals is effected by 
 pouring superheated thermit steel around the parts to be united. 
 Thermit is a mixture of finely divided aluminium and oxide. This 
 mixture is placed in a crucible, and the steel is produced by igniting 
 the thermit in one spot by means of a special powder, which gener- 
 ates the intense heat necessary to start the chemical reaction. When 
 the reaction is once started it continues throughout the whole mass, 
 the oxygen of the iron being taken up by the aluminium (which has 
 a strong affinity for it), producing aluminium oxide (or slag) and 
 
 Fig. 83. — The Above is a Thermit Weld, in which New Teeth have been 
 
 Welded in. 
 
 superheated thermit steel. Ordinarily, the reaction requires from 
 thirty-five seconds to one minute, depending upon the amount of 
 thermit used. As soon as it ceases, the steel sinks to the bottom 
 of the crucible, and is tapped into a mould surrounding the parts to 
 be welded. As the temperature of the steel is about 5,400° F. it 
 fuses and amalgamates with the broken sections, thus forming a 
 homogeneous weld. 
 
 It is necessary to preheat the sections to be welded before pour- 
 ing to prevent the chilling of the steel. The principal steps of the 
 operation are : to clean the sections to be welded ; to remove enough 
 metal at the fracture to provide for a free flow of thermit steel ; to 
 align the broken members and surround them with a mould to retain 
 
 154 
 
THERMIT WELDING 
 
 155 
 
 the steel ; to preheat by a torch or other suitable heater to prevent 
 chilling the steel; to ignite the thermit and tap the molten steel 
 into the mould. 
 
 This process is specially applicable to the welding of large sections. 
 It has been extensively used for welding locomotive frames, broken 
 motor castings, rudders and sternposts of ships, crankshafts, spokes 
 of driving wheels, connecting-rods, and heavy repair work in general. 
 One great advantage of the thermit process is that broken parts can 
 usually be welded in place. For example, locomotive frames are 
 welded by simply removing parts that would interfere with the appli- 
 cation of a suitable mould. Thermit is also used for pipe welding 
 and in foundry practice to prevent the " piping " of ingots. 
 
 Fig. 84. — Thermit- Welded Crosshead. 
 
 Preparation. — The first step in the operation of thermit welding 
 is to clean the fractured parts and cut away enough metal to ensure 
 an unobstructed flow of the molten thermit. The oxy-acetylene 
 cutting blowpipe is very efficient for this operation. The amount 
 that should be cut away depends upon the size of the work. Assum- 
 ing that a locomotive frame is to be welded, the space should be 
 about £ inch wide for a small frame, and 1 inch wide for a large 
 frame. The frame sections are then jacked apart about f inch to 
 allow for contraction of the weld when cooling. Trammel marks 
 are scribed on each side of the fracture to show the normal length. 
 If the weld is to be made on one member of a double-bar frame, 
 the other parallel member should be heated with a blowpipe to 
 equalise the expansion in both sections and prevent unequal 
 strains. 
 
 Fig. 84 shows a Thermit-Welded Crosshead, which was broken 
 
156 MODERN METHODS OF WELDING 
 
 through the middle before welding took place. The two pieces of the 
 crosshead, before welding, are bevelled on each edge of the fracture, 
 so as to give a further area of thermit, thereby getting greater 
 strength in the weld. The bevelling was done with an oxy-acetylene 
 cutter, using acetylene and oxygen. 
 
 Mould for Thermit W elding '.—The mould surrounding the frac- 
 tured part should be so arranged that the molten thermit will run 
 through the gate to the lowest part of the mould and rise through 
 and around the parts to be welded. The thermit steel is poured 
 through the gate and forms a riser which rises into a space after 
 passing round and between the ends of the fractured crosshead. 
 The thickest part is directly over the fracture, and the band overlaps 
 the edges of the fracture by at least one inch. An opening is also 
 made for preheating the ends to be welded. 
 
 Patterns for the riser, pourings, and heating gates can be made 
 of wood. The riser should be quite large enough, because the steel 
 that first enters the mould is chilled somewhat by coming in con- 
 tact with the metal even when preheated. This chilling effect is 
 overcome by using enough thermit steel to force the chilled portion 
 into the riser and replacing it by metal which has practically the 
 full temperature received during reaction. When the mould and 
 the box are filled and tamped, the wooden runner and riser patterns 
 are withdrawn. The mould is then ready for the preheating and 
 the drying operation, which causes the wax matrix to melt and run 
 out. The mould must be made of some refractory material, owing 
 to the intense heat. 
 
 Thermit welding was first introduced in 1903 and was adopted 
 on marine work, which has had a great many successful welds of this 
 nature. It is being used largely by railway and tramway companies. 
 One can, nearly always, see the process being worked in the streets 
 on the tramway lines. 
 
 There are many technical schools in every large city, where in- 
 struction and practice are given in thermit welding, as well as other 
 welding processes. 
 
 Thermit Required for Welding.- — The quantity of thermit re- 
 quired for making a weld can be determined from the cubic content 
 of the weld. Calculate the contents of the weld and its reinforce- 
 ment in cubic inches, double this amount to allow for filling the gate 
 and riser, and multiply by 0-56 to get the number of pounds of 
 thermit required. When wax is used for filling, the weight of the 
 thermit can be determined as follows : Weigh the wax supply before 
 and after filling the fracture. The difference in weight (in pounds) of 
 
THERMIT WELDING 157 
 
 the quantity used multiplied by 22 will give the weight of thermit 
 in pounds. 
 
 When a quantity of more than 10 pounds of thermit is to be used, 
 add 10 per cent, of steel punchings (not over J inch diameter), or 
 steel scrap, free from grease, to the thermit powder. If the thermit 
 exceeds 50 pounds, 15 per cent, of small mild steel rivets may be 
 mixed with it. One per cent, by weight of pure manganese and 
 1 per cent, of nickel thermit should be added to increase the strength 
 of the thermit steel. 
 
 Preheating — Making a Weld. — The ends to be welded should be 
 red-hot at the moment the thermit steel is tapped into the mould. 
 This preheating is done preferably by a gasolene compressed air 
 burner. As previously mentioned, it melts the wax matrix used for 
 filling the fractures to form the pattern for the reinforcing band. 
 When the ends have been heated red, quickly remove the burner and 
 plug the preheating hole with a dry sand core, backing it up with a 
 few shovelfuls of sand, well packed. The end of the cone-shaped 
 crucible should be directly over the pouring gate and not more than 
 4 inches above it. To start reaction, place J teaspoonful of ignition 
 powder on the top of the thermit and ignite with a storm-match. 
 It is important that sufficient time be allowed for the completion of 
 the thermit reaction and for the fusion of the steel punchings which 
 have been mixed with the thermit. 
 
 With charges containing from 30 to 40 pounds of thermit, the 
 crucible should not be tapped in less than thirty-five seconds; 
 with charges containing 50 to 75 pounds, forty seconds; 75 to 100 
 pounds, fifty seconds to one minute. When welding a broken 
 frame, as shown previously, the screw jack used for forcing apart 
 should be turned back somewhat to relieve the pressure gradually 
 as the weld cools. After pouring the mould should remain in place 
 as long as possible (preferably ten to twelve hours) to anneal the 
 steel in the weld; and, in any case, it should not be disturbed at 
 least two hours after pouring. When welding a broken spoke in a 
 driving wheel or a similar part, it is necessary to preheat the adjacent 
 spokes in order to prevent undue strains through expansion and 
 contraction. If a section of a spoke is broken out, it can be cast in, 
 but if the space is over 6 inches long it is better to insert a piece of 
 steel and make a weld at each end. Owing to the high temperature 
 (5,400° F.), and the violent ebullition of thermit during reaction, 
 the crucible must be relined with a very refractory material. The 
 crucibles used for this purpose have sheet-iron shell and are lined 
 with magnesia. 
 
158 
 
 MODERN METHODS OF WELDING 
 
 Filling Shrinkage Holes and Surface Flaws. — The filling of surface 
 flaws in castings and forgings usually requires from 2 to 10 pounds of 
 thermit. 
 
 To make a weld of this kind, place an open mould around the 
 part to be filled large enough to overlap it about § inch ; clean the 
 hole thoroughly and heat to red-heat by means of a strong blow- 
 burner. Use 18 ounces of thermit for each cubic inch of space, 
 
 Fig. 85. — Thermit- Welded Rock Crusher. 
 
 but not less than 2 pounds for any one weld. Place a small amount 
 of thermit in the crucible, which, in this case, is of a small size for 
 hand use. Ignite the thermit with the ignition powder, and as 
 soon as it begins to turn add the remainder, feeding it fast enough 
 to keep the combustion going. When the reaction is completed 
 quickly pour the slag (which is about three-fourths of the liquid) 
 into dry sand. Then pour the steel into the open mould and sprinkle 
 
THERMIT WELDING 159 
 
 loose thermit on the top to prolong the reaction, as the casting, even 
 when preheated, will have a chilling effect on the steel. 
 
 Composition of Thermit Steel. — An average analysis of thermit 
 steel is as follows: Carbon, 0-05 to 0-10; manganese, 0-08 to 0-10; 
 silicon, 0-04 to 0-05; aluminium, 0-07 to 0-18 per cent. The tensile 
 strength is about 65,000 pounds per square inch. 
 
 Fig. 85 is a remarkable weld of a rock crusher for the Casparis 
 Stone Company. The eccentric bearing broke off, leaving a 
 fractured surface extending longitudinally through the bearing, 
 measuring 6 feet 2 inches long and an average of about 7 inches 
 thick. A mechanical repair was first attempted on the new brake, 
 which, however, failed only after a few days' service. Resort was 
 then made to the thermit process, and the casting was shipped 
 to the thermit factory. The broken sections were lined up, a gap 
 of about 3 inches between the sections was cut out with the oxy- 
 acetylene flame, 90 pounds of wax applied in the welding gap, and 
 a mould box built round the fracture. Six preheating gates were 
 made in the mould. The preheating was begun at 4 a.m. on the day 
 of the pour; 1|- hours of preheating were required to burn all the 
 wax out of the mould. The preheating was kept up for about twelve 
 hours until 3.45 p.m., the time the reaction started. The weld is 
 illustrated on p. 158. 
 
 Particular interest attaches to this repair because an extra large 
 crucible, having a capacity of 2,000 pounds of thermit, and of some- 
 what different design from the standard crucible, was used experi- 
 mentally. The crucible was filled almost to its 2,000 pounds 
 capacity, 1,750 pounds of railroad thermit being used. In spite of 
 the enormous amount of thermit formed in one crucible, the re- 
 action and pour were accomplished Avith entire success. The crucible 
 rested steadily and motionless on its four supports throughout the 
 reaction. Sixty seconds were allowed for the reaction to take place, 
 after which the crucible was tapped in the usual manner, in the case 
 of large welds by means of a long iron rod. From the moment of 
 tapping it took two and three-quarter minutes for the contents to 
 run out, as compared with the one minute generally required in the 
 case of number 10 crucible. The illustration is self-explanatory. 
 It shows the arrangement of the gates and riser, and indicates the 
 excess of metal present after the mould box was finally removed. 
 
CHAPTER XXV 
 PROPERTIES OF PRINCIPAL NON-FERROUS METALS 
 
 Scattered throughout the pages of technical literature are various 
 references to non-ferrous metals and alloys, the importance of which 
 is apt to be lost sight of because they become inaccessible after a 
 short time. It is therefore desirable that such information should be 
 carefully sifted and what is useful in it collated and presented in a 
 handy form. Such is the purport of the present chapter. 
 
 The science of metallurgy has developed wonderfully within the 
 last few years, especially with regard to the non-ferrous metals. 
 Manufacturers are awakening to the fact that many of the disturbing 
 influences which mar their best efforts are due to prevalent miscon- 
 ceptions respecting the combined chemical compositions and the 
 physical structures of the materials, and that henceforward science 
 and practice must go hand in hand if true progress is to be attained. 
 
 An ordinary chemical analysis, supplemented by the usual 
 physical tests, was, at one time, considered to give the total history 
 of an alloy. Things have changed, however, and it is now recognised 
 that metals and compounds may be incorporated in an alloy under 
 conditions which would so change the arrangements of the con- 
 stituents as to render it difficult, if not impossible, to determine 
 the original state of combination or the ultimate condition of the 
 product. There has been no lack of fanciful theories in regard to 
 the segregation, crystallisation, and fatigue of metals, some of 
 them based on insufficient data derived from purely physical 
 and chemical tests. 
 
 A new branch of metallurgical study has recently come into 
 prominence under the name of "metallography " — that is, the micro- 
 scopical examination of the structure of metals, which has already 
 been the means of revealing the causes of many peculiarities of metals 
 and their alloys and confirming other theories concerning them 
 which used to be looked upon as parabolic. One of the most im- 
 portant properties or changes which occur in alloys is that of liquida- 
 tion, which was only proved by metallographic examination. 
 
 When a solution fluid at ordinary temperature is allowed to cool 
 
 160 
 
PROPERTIES OF PRINCIPAL NON-FERROUS METALS 161 
 
 below its congealing-point, the process frequently takes place in such 
 a manner that, as cooling progresses, certain constituents of the 
 solution congeal first, whilst the solution still remaining liquid under- 
 goes constant changes in composition until a certain point is reached, 
 after which this solution also congeals. The solution congealing last 
 is called the eutectic (most fluid) solution. On examination of the 
 latter, it will be found that during cooling a disintegration of the 
 constituents previously dissolved in one and another has taken place, 
 and that the solution now forms only an intimate mixture of these 
 constituents. 
 
 Many alloys show a similar behaviour when cooling. If, for 
 instance, a melted zinc-copper alloy containing more than 72 per 
 cent, of zinc be allowed to cool, zinc crystals are first separated, 
 while an alloy poorer in zinc still remains liquid. This separation 
 of zinc is continued until the content of the zinc has been reduced 
 to 72 per cent., which takes place when the temperature has fallen 
 to 1,404° F. This is the eutectic point: a eutectic alloy which no 
 longer separates any constituents but solidifies throughout at that 
 temperature consists, therefore, of 72 parts of zinc and 28 parts of 
 copper. In congealing it disintegrates, however, to an intimate 
 mixture of its constituents which, on reheating, first dissolve again 
 in one another, and with an increase of temperature gradually dis- 
 solve the previously separated zinc. If, on the other hand, the alloy 
 contains less than 72 per cent, of zinc and more than 28 per cent, 
 of copper, copper is, in congealing, first separated till, at 1,404° F., 
 the composition of the eutectic alloy has again been reached and 
 then also congeals. 
 
 Specific Gravity. — The specific gravity or density of alloys corre- 
 sponds only in a few cases with that which would result by calcula- 
 tion from the specific gravities of the constituents. The specific 
 gravity should be calculated from the volumes and not from the 
 weights. Dr. Ure gives the correct rule as follows : Multiply the sum 
 of the weight into the products of the two specific gravity numbers 
 for a numerator and multiply each specific gravity number into the 
 weight of the other body, and add the products for a denominator. 
 The quotient obtained by dividing the said numerator by the 
 denominator is the computed mean specific gravity of the alloy. 
 With regard to the influences exerted upon the strength of metal 
 by alloying, the following general law may be laid down. By the 
 absorption of a foreign body the strength of the metal is increased. 
 It grows with the content of the foreign body until the latter has 
 reached a certain proportion, which varies in individual cases. 
 
 11 
 
162 
 
 MODERN METHODS OF WELDING 
 
 When this limit has been passed, the strength again decreases, fre- 
 quently with great rapidity, provided that the body itself does not 
 possess greater strength than the metal. • By the addition of a third 
 metal to an alloy consisting of two metals, it is sometimes possible 
 to bring about an additional increase in strength. 
 
 Limit of Elasticity. — In alloying a metal the limit of elasticity 
 increases steadily with the breaking strength, limit of elasticity, and 
 breaking weight moving more closely together. The limit of elasti- 
 city usually increases still further when, with the increase of the 
 foreign body added, the highest degree of strength has already been 
 attained, and a decrease in strength reappears. Limit of elasticity 
 and strength sometimes finally converge. 
 
 Mixtures of Alloys. 
 
 
 Copper. 
 
 Tin. 
 
 Zinc. 
 
 
 Bell-metal ! 78 
 
 22 
 
 
 Standard bell-metal 
 
 Gun-metal 
 
 90 
 
 10 
 
 — 
 
 Ordnance castings 
 
 Gun-metal 
 
 
 88 
 
 10 
 
 — 
 
 Steam chest pumps 
 
 Gun-metal 
 
 
 86 
 
 14 
 
 ■ — 
 
 Hard bearing metal 
 
 Naval brass 
 
 
 62 
 
 1 
 
 37 
 
 Stanchions, tube plates 
 
 Sheet brass 
 
 
 70 
 
 — 
 
 30 
 
 For sheet tubes 
 
 Ordinary brass 
 
 
 66f 
 
 — 
 
 33* 
 
 General use 
 
 Manganese bronze 
 
 
 56 
 
 0-9 
 
 41 
 
 
 Gun-metal ord. 
 
 
 88 
 
 8 
 
 2 
 
 General use 
 
 Yellow brass 
 
 
 85 
 
 3 
 
 15 
 
 General plumbing work 
 
 
 
 
 Iron. 
 
 
 Phosphor-bronze 
 
 89-5 
 
 10 
 
 0-5 
 
 Heavy bearings 
 
 
 
 
 
 Mdting-Points. 
 
 Boiling -Points. 
 
 
 Degrees C. 
 
 Degrees G. 
 
 Aluminium 
 
 658-7 
 
 1,800 
 
 Copper 
 
 
 
 
 1,803 
 
 2,310 
 
 Iron 
 
 
 
 
 1,520 
 
 2,450 
 
 Tin 
 
 
 
 
 231-9 
 
 2,270 
 
 Zinc 
 
 
 
 
 419-4 
 
 905-7 
 
 Gun-metal 
 
 
 
 
 995 
 
 1,825 
 
 Red brass 
 
 
 
 
 970 
 
 1,780 
 
 Low-grade brass . . 
 
 
 
 
 980 
 
 1,795 
 
 Bronze with zinc 
 
 
 
 
 980 
 
 1,795 
 
 Cast yellow brass 
 
 
 
 
 895 
 
 1,645 
 
 Naval brass 
 
 
 
 
 855 
 
 1,520 
 
 Manganese bronze 
 
 
 
 
 870 
 
 1,600 
 
PROPERTIES OF PRINCIPAL NON-FERROUS METALS 163 
 
 The purest metals possess the greatest flexibility. By alloying 
 this property is diminished, and sometimes almost reduced to 
 nothing. The melting temperature of metals is frequently lowered 
 by alloying. Therefore it is essential that, when welding takes place 
 upon non-ferrous metals, precautions must be taken not to add 
 impure metals from the welding -rod into the welded portion. 
 Also it is important that the metal article being welded and the 
 welding-rod shall be one and the same material, plus ingredients to 
 replace the element that is volatilised during welding. 
 
CHAPTER XXVI 
 DELTA METALS 
 
 The alloys known under the name of " delta metals " are a series 
 of high-class engineering alloys, of which the first was placed on 
 the market in 1885 by the eminent metallurgist, the late Alexander 
 Dick. These metals have been greatly developed and comprise a 
 whole group of different alloys. Hence there are naturally always 
 welding repairs in these metals to be done. A description of their 
 properties will enable the student to distinguish them from other 
 metals and alloys, so that the treatment may be administered as 
 required. 
 
 It is obviously impossible to combine in one single alloy all the 
 physical and chemical properties suited to a great variety of pur- 
 poses. The different standard alloys vary in composition accord- 
 ing to the purposes for which they are more particularly adapted. 
 
 Some alloys possess in every degree the properties of malleability, 
 strength, and resistance to corrosion; others are superior bearing 
 metals. Some have particular qualifications for electrical purposes ; 
 others for high-speed machining for brass-founder's work. One 
 is called silver bronze, possessing a silver-white colour. 
 
 Metal No. 1 : Strongest malleable bronze for high-tensile forgings, 
 castings, and rods. 
 
 Metal No. 2 : Silver bronze (improved nickel silver) rods, forgings, 
 and castings. 
 
 Metal No. 3 : Specially adapted for solid drawn tubes. 
 
 Metal No. 4: (Various grades) malleable bronze, strong as steel, 
 tough as wrought iron, highest resistance to corrosion, for castings, 
 forgings, stampings, rods, sheet, wire, etc. 
 
 Metal No. 5: Antifriction bronze for bearing castings. 
 
 Metal No. 6: Improved gun-metal for castings of every descrip- 
 tion. 
 
 Metal No. 7: Bronze to resist high temperature, castings, forg- 
 ings, stampings, rods, etc. 
 
 Metals Nos. 8, 9, 9a: Various grades of white antifriction metals. 
 
 Of the various alloys, the most used is No. 4. Its great strength, 
 
 164 
 
DELTA METALS 165 
 
 equalling that of steel, its elongation, its toughness, its malleability, 
 and its property of resisting in a marked degree the corrosive action 
 of sea and mine water, chemicals, gases, etc., render this particular 
 brand the most useful for all classes of work in which durability, 
 strength, and reliability are the qualities chiefly to be taken into 
 consideration. It should be borne in mind that when objects made 
 from different metals or alloys are, while in metallic contact with 
 each other, immersed in sea water, brine, or any other exciting fluid, 
 galvanic action will be set up, which will bring about deterioration 
 or decomposition of the metals. The rapidity with which this 
 deterioration takes place, and also the question of which of the metals 
 is chiefly attacked, depends upon the relative position of the different 
 metals in the electric scale. Metals hardly suffer at all when in con- 
 tact with those which are electro-negative to them ; but when in 
 contact with a metal which is electro-positive towards them they 
 are rapidly destroyed. For this reason delta metal should never be 
 placed in metallic contact with copper or gun-metal when immersed 
 in sea water, or used in running machinery or other plant which is 
 exposed to the action of corrosive fluids. 
 
 From this it is clear that operators should take care in the weld- 
 ing of delta-metal articles. A welding-rod must be used of the same 
 constituents as the article being welded, plus the deoxidising sub- 
 stance. The results of tests from this metal have shown it to have 
 a tensile strength of 24 tons per square inch, elongation from 30 to 
 40 per cent., and limit of elasticity 19-02 tons per square inch. 
 When reheated to 550° C. (a dull red colour) it becomes soft, and 
 is then one of the most malleable copper alloys in existence, as it is 
 in a semiplastic state, in which it can be worked as easily as wrought 
 iron, and can be stamped, forged, and pressed to any extent required. 
 As these operations add 50 per cent, strength to the metal, without 
 impairing any of its other valuable qualities, it is obvious that, for 
 the majority of uses, the wrought material is to be preferred, the 
 more so as it is free from defects which are sometimes found in 
 castings, such as blowholes, etc. 
 
 Forged bars of this alloy show, as a result of four tests, a tensile 
 breaking strain of 34-4 tons per square inch, with an elongation of 
 26-25 per cent. Its great strength is but little affected by increase 
 of temperature. This quality adds considerably to its value for 
 engineering purposes, such as engine fittings exposed to hot steam. 
 The result of the test at a temperature of 506° F. was that it lost 
 only about 17| per cent., while at 500° F. brass lost 38 per cent., 
 phosphor-bronze about 31 per cent., and gun-metal 33 per cent. Delta 
 
166 MODERN METHODS OF WELDING 
 
 metal can be roughly described as a copper-zinc alloy, chemically 
 combined with definite proportions of iron and other elements. 
 The secret lies, not only in the use of virgin metals in the exact pro- 
 portions, but still more in the proper methods of combining these, 
 and eliminating during the manufacturing processes certain other 
 elements after these have produced the desired effect. 
 
 The foregoing description is one which sets out the properties 
 of a metal very largely used in many workshops. The author 
 has not seen them detailed before; but he has had considerable 
 experience in the welding of it, and articles made of it are now 
 coming into various workshops or welding depots for repairs. 
 
 Welding-Rods. 
 
 Welding-rods for use with delta-metal articles should contain the 
 same pure metal, and also a trace of phosphorus and aluminium. 
 These are added to the metal to prevent oxidation. The welding-rod 
 should be manufactured from pure metal, and the constituents or 
 ingredients uniformly distributed through the mass. They should 
 be made in sizes from ^ to \ inch diameter and 24 inches long. 
 They are usually drawn down into wire of various thicknesses and 
 stocked by the manufacturers. They should also be made in various 
 grades to suit the articles to be welded. 
 
 Preparation of Articles. 
 
 The welding of delta metal is easy of application, as the metal 
 is very pure. The area to be welded must be bevelled in the usual 
 course. If it is not over § inch thick it is only necessary to bevel 
 one side, but if over f inch thick it is necessary to bevel both sides. 
 Afterwards the weld has to be thoroughly cleaned. This is very 
 important, because the molten metal will not adhere to a greasy 
 surface. 
 
 Preheating. — The laws of expansion and contraction have neces- 
 sarily to be considered in welding this metal. Therefore, it is desir- 
 able always to preheat the articles, and, after welding sharply, they 
 should be returned to an annealing furnace to heat up and allowed 
 to cool slowly until quite cold. Apart from the question of expan- 
 sion of the metal, preheating saves a good quantity of gases in 
 welding. 
 
 Blowpipe Tower. — The power of the blowpipe is, generally speak- 
 ing, the same as for copper, or a size larger than that used for iron 
 or steel. The regulation of the flame is very important. This must 
 
DELTA METALS 167 
 
 be done with great accuracy, and there must be no excess of acety- 
 lene or oxygen. The former would carbonise the weld, the latter 
 would oxidise it. The oxygen pressure must not at all be increased 
 over that stated by the makers. The flame should be regulated 
 till a clear white jet or cone is perceived and kept at this until the 
 blowpipe gets somewhat heated and the flame begins to get less, 
 when a little more acetylene should be turned on to make the even 
 cone. In no case should excess oxygen be turned on. 
 
 Method of Welding. — The preheated and bevelled article should 
 be on the welding table, the blowpipe regulated, and the weldinr- 
 rod in the left hand. Approach the welding line at about \ inch 
 from the edge, keeping the white top of the flame -£% inch from the 
 metal. As the melting starts, the blowpipe should be passed over 
 to the edge, and this melted. At this period the welding-rod should 
 be nearly at the welding-point. As soon as the edge is melted, put 
 a little from the welding-rod into the molten mass to fill up the 
 bevel, and continue the movement of the blowpipe along the line 
 of welding in a gyratory movement, advancing at the same time 
 with the necessary addition of the welding-rod to fill up the spaces 
 until there is enough metal added to fill up the bevel. The welding 
 must be continuous when once started, and one must not go over 
 the weld twice without adding additional welding-rod. The flux 
 must be used along with the welding-rod, and dipped while hot into 
 the flux jar. No more must be used than is necessary — just sufficient 
 to clean the metal and prevent oxidation. When the weld has been 
 completed, the welded article should be put back into the annealing 
 furnace and heated to 550° C, then, when cold, the weld may be 
 hammered. 
 
 Failures. — These only occur through lack of metallurgical know- 
 ledge: adding a rod of inferior quality; bad penetration; allowing 
 the oxide to form internally in the metal, forming blowholes ; being 
 too long on the weld, and causing the metal to become too liquid, 
 burnt, and oxidised ; or melting the metal to its boiling-point instead 
 of its melting-point. 
 
 All these defects are easily overcome. They require a little 
 practice and careful study of the metallurgical and technical points. 
 With this knowledge failure is impossible. 
 
CHAPTER XXVII 
 ALUMINIUM 
 
 Aluminium has a silvery-white appearance, and is capable of taking 
 a very high polish. It is one of the soft metals, its hardness being 
 only about 2-5. Its most valuable property is its lightness, the 
 specific gravity being 2-56, varying slightly with the impurities 
 present. Rolling, hammering, and stamping increase its specific 
 gravity somewhat, that of worked metal being as high as 2-7. 
 The tensile strength of aluminium castings is about 6 or 7 tons per 
 square inch in section, although wire may reach 15 to 30 tons> 
 depending on its fineness. 
 
 The elastic • limit is about 3 to 4 tons in the case of castings, 
 and may be as high as 15 tons for wire. The metal flows readily 
 under pressure, and is therefore both malleable and ductile. It can be 
 rolled into very thin sheets or drawn into fine wire. Aluminium 
 melts readily, its melting-point being 658° C, and its boiling-point 
 is estimated at 1,800° C, being non- volatile in ordinary circum- 
 stances. It has a very specific heat, about 0-308, which remains 
 constant up to about 800° C. Its latent heat of fusion is given as 
 100. It is a good conductor of electricity, the conducting power 
 of pure aluminium (silver taken as 100) being about 56; but this 
 should be modified by the presence of impurities, and by the condi- 
 tion of the metal, whether annealed or not, owing to its lightness. 
 For equal conductiveness the weight of the aluminium is about 
 48-5 per cent, that of copper. It is used largely in the manu- 
 facture of alloys and for many purposes, and is in big demand for 
 motor-car castings, whilst its further use has been extended by 
 the advent of the aeroplane. Owing to its low tensile strength, its 
 usefulness has been curtailed for many engineering purposes. 
 Hence many aluminium alloys have been brought out in the effort 
 to combine strength with lightness. 
 
 The lightness, low cost, and ease of working peculiar to sheet 
 aluminium have combined to make it one of the most popular metals 
 for the manufacture of various articles from the sheet form. The metal 
 can be obtained in grades from dead soft to hard rolled. A square 
 
 168 
 
ALUMINIUM 169 
 
 foot of 14 S.W.G. sheet aluminium weighs 1-11 pounds, the same 
 size copper weighs 3-70, and brass 3-56 pounds. The difference in 
 prices for the same sizes is as follows : 
 
 One square foot of sheet aluminium costs Is. 2Jd. 
 „ copper „ 3s. 
 
 ,, ., ,, brass ,, 2s. 6d. 
 
 Hence aluminium is much less than half the cost of other non- 
 ferrous metals ; in sheet form it is undoubtedly (with the exception 
 of iron) the cheapest material on the market. For this reason sheet 
 aluminium has come into extended use for such work as motor 
 body and railway coach construction, for ceilings and panels, 
 and many other purposes. The welding of this sheet aluminium 
 is spreading rapidly. Operators should devote much time to its 
 study from a metallurgical point of view, so that when they come 
 to weld it they may understand what takes place when the heat 
 of the flame is used on the sheet and it becomes molten. 
 
 Aluminium is without doubt the most difficult to weld of all 
 metals. This is largely due to the difference in the fusibility of 
 the aluminium oxide and aluminium metal. When two separate 
 pieces of aluminium are welded together at their edges, by means of 
 the welding flame, the melted parts do not flow properly together, 
 as is the case with iron, where the melting-point of the oxide is lower 
 than that of the metal. At high temperatures aluminium has great 
 affinity for oxygen. The molten parts become covered, under the 
 influence of the welding flame, with a fine coating of oxide, which has 
 great power of resistance to the flame, and, on cooling, the parts 
 remain unjoined. Therefore, if aluminium parts are to be welded 
 together properly, this skin of oxide must in some way be destroyed. 
 This can be done to some extent mechanically. The destruction of 
 the covering of the oxide can be brought about by moving or puddling 
 the molten metal of the weld by means of the aluminium wire used 
 as a feeding-rod to let the separate drops, already formed, flow to one 
 another. With this method, however, there is a danger that the 
 weld will be a failure, not a homogeneous one. Almost certainly soma 
 part or other of the oxide will remain in the weld, which would prob- 
 ably make it defective. There are also many points to watch care- 
 fully in the welding of aluminium. It is imperative that it be sup- 
 ported on the underside of the weld. Otherwise, as soon as the 
 melting takes place the molten metal would fall through, leaving a 
 hole in the weld, which the student would find it difficult to fill up 
 without causing further holes and burning the aluminium, causing 
 
170 MODERN METHODS OF WELDING 
 
 even further oxidation. Mechanical puddling should not be prac- 
 tised, because it is only a makeshift, which is seldom successful. 
 Proper welding — that is, a homogeneous joint — can only be done 
 by very careful study and practice. An important point is the 
 use of a good flux, which will cause the oxide to melt or break 
 down to the same temperature and at the same time as the metal 
 itself. 
 
 The difficulty will be seen when it is explained that the melting- 
 point of metallic aluminium is 650° C, whilst the melting-point of 
 aluminium oxide is 1,800° C. To produce a flux that will dissolve 
 the oxide at the low melting-point of the metal, and at the same time 
 protect the hot metal from contact with the air, is no easy problem 
 to the chemist and metallurgist. It is only recently that such fluxes 
 have become obtainable. These vary considerably in their elements 
 and compositions. There are several at present being marketed. 
 Each one pleads it is the best; some are good, but others are no use 
 whatever. If a good flux is obtained, there should be no reason 
 (after constant practice) why operators should not make very satis- 
 factory welds in aluminium. Light hammering of the weld and 
 reheating to a temperature of about 450° C. are beneficial. Alumin- 
 ium welding is now quite an important branch of the oxy-acetylene 
 welding industry, and is employed more particularly in connection 
 with brewing, where aluminium is largely superseding enamelled 
 ware. Fluxes should always be removed by washing off as the 
 welding job is complete. 
 
 It is very important to choose with care a correct blowpipe for 
 the welding of aluminium sheeting. This must be a very light one, 
 much lighter than for the same thickness of iron or steel: just 
 half the power would be plenty. One must also be very careful 
 to watch the oxygen pressure. This must be less than that specified 
 by the makers, which is based on iron and steel. In all cases of 
 welding pure aluminium it is imperative that the edges to be welded 
 shall be absolutely clean, and the welding-rod of the purest metal 
 obtainable, so as not to get impurities into the weld, which would 
 cause it to be defective. 
 
 Aluminium being of low melting-point, the operator requires 
 great patience and skill when welding, and must avoid burning the 
 metal, or getting much above its melting-point, as the heat spreads 
 rapidly, especially on light work. 
 
 If the metal melts too much on each side of the weld and makes 
 too large a molten bath, the result is a rough and probably defec- 
 tive weld. Therefore, neither the blowpipe nor the flame must be 
 
ALUMINIUM 
 
 171 
 
 large, and the oxygen pressure must be just sufficient to keep in the 
 flame, which must have the smallest jet possible. 
 
 The welding-rod for aluminium welding is usually made from 
 pure aluminium drawn wire of diameters for the various thicknesses 
 to be welded. These rods are usually kept in stock by various 
 factors of acetylene equipment. 
 
 It is imperative that the rods be pure, and free from even a 
 trace of copper. Copper is very detrimental to welds and causes 
 (in moisture or water) corrosion. In welding pure aluminium parts 
 it is not always necessary to preheat, because in many cases the 
 
 Fig. 86. — Fractured Aluminium Gear Case. 
 
 aluminium article is able to stand the welding without preheating ; 
 but hammering and annealing afterwards increase the strength of 
 the weld greatly. If flux has been used, the article should be brushed 
 in running water if possible, because the flux has a corroding effect 
 on the metal. 
 
 The above illustration shows a motor aluminium crank case 
 which was welded successfully and afterwards annealed. A neat 
 job was made and in no way distorted. 
 
 In dealing with alloys the most important properties which are 
 desired are strength and durability combined with ductility. Tests 
 have been applied to commercial specimens of alloys upon the market. 
 
172 
 
 MODERN METHODS OF WELDING 
 
 Alloys which are lighter than aluminium itself generally contain 
 magnesium, which reduces its tensile strength, and renders it 
 brittle and less permanent than aluminium itself. An alloy con- 
 taining approximately 76 per cent, of aluminium, 21 per cent, of 
 zinc, 3 per cent, of copper, has an ultimate strength of 12 tons per 
 square inch. This is a mixture from which motor crank cases are 
 frequently made. 
 
 Copper alone does not result in any great gain in beneficial proper- 
 ties, as is seen by the fact that an alloy approximately 95 per cent . 
 aluminium and 5 per cent, copper only reached a maximum ultimate 
 stress of 8 tons per square inch. With the exception of duralumin, 
 none of the commercial alloys investigated showed any remarkable 
 excellence, or, indeed, bore out the claims of the makers. Indepen- 
 dent investigation in the laboratory with alloys of definite com- 
 position afforded interesting results which were exhaustively tabu- 
 lated according to the method of mechanical and heat treatment. 
 The result of annealing after treatment was invariably to lower the 
 ultimate strength, while rolling had the opposite effect. 
 
 Sand cast alloy containing 15 per cent, of zinc had an ultimate 
 strength of 11-19 tons per square inch, but a rolled bar reached in 
 one case to 17 tons, wire being 19 tons per square inch, even after 
 annealing at 400° C. An alloy containing 20 per cent, zinc showed 
 higher tensile strength all round, sand cast 17 tons per square inch, 
 1 \ inch diameter rolled bar 22J tons per square inch. Copper alone 
 in small quantities does not cause any appreciable improvement, 
 but it can be advantageously used in conjunction with zinc. Experi- 
 ments with aluminium- zinc alloys to which 3 per cent, of copper 
 was added showed ultimate strength as follows : 
 
 15 per cent. Zn 
 
 20 
 
 26 
 
 Sand Cast. 
 
 14-15 tons 
 15-55 „ 
 
 18-25 „ 
 
 Chill Cast. 
 
 14-9 tons 
 14-2 
 22-22 ,, 
 
 Rolled Bar. 
 
 23-6 tons 
 23 
 
 27-92 „ 
 
 The effect of magnesium in small quantities is, on the other hand, 
 most decided, and from 25 to 5 per cent, magnesium has resulted in 
 an alloy which, in rolled condition, possesses an ultimate strength of 
 28 tons per square inch. Light alloys are fairly permanent, if not 
 exposed to high temperatures, although cases of deterioration have 
 been known with alloys of approximately 80 per cent, zinc and 20 per 
 cent, aluminium. The chief trouble is corrosion, which is marked in 
 
ALUMINIUM 173 
 
 the case of alloys containing copper. Indeed, the corrosion of all light 
 alloys is hastened by contact with copper or brass when immersed. 
 The use of light alloys in constructional work often entails welding, 
 etc., and great care should be exercised, as these alloys are generally 
 very sensitive to all such treatment, which may thus lead to an un- 
 expected failure. 
 
 Certain aluminium alloys, generally known as duralumin, 
 became materials of high importance during the war, and owe their 
 great development to their mechanical properties. Some of these 
 are singular and due apparently to method of tempering. Investi- 
 gations were made with a duralumin of the following composition : 
 Aluminium 93-9, magnesium 0-43, copper 3-7, manganese 0-Q, 
 zinc 0-25, silicon 0-58, iron 0-53 per cent, (some of these minor con- 
 stituents may probably be regarded as accidental). Following 
 the ordinary practice, the metal was heated to 450° C, quenched in 
 cold water, and left to itself. The quenching itself did not seem 
 to change the properties of the alloy to any important extent, but 
 the breaking strength, impact strength, elastic limit, and hardness 
 increased afterwards, within a day or two, while the elongation and 
 reduction in area were little affected. 
 
 Aluminium alloy that is to be welded should be scraped and 
 cleaned, and if the stock is more than \ inch thick the edges should 
 be bevelled. If the blowpipe, when welding, appears too fierce 
 a flame, then this must be reduced by (1) reduction of oxygen, and 
 (2) an excess of acetylene. This excess of acetylene does not injure 
 aluminium alloy, but lowers the flame temperature, which is desirable, 
 owing to the low melting-point. 
 
 Coal-gas, instead of acetylene, mixed with oxygen would do for 
 welding aluminium, as it is a softer flame. Often good sound welds 
 are made with these gases, and it is very easy to fix up to the town 
 gas; one precaution must be taken — that is, the coal-gas must be 
 passed through an hydraulic safety valve the same as acetylene; 
 the coal-gas is under the same pressure as the acetylene, therefore 
 it has to be used in the same way. Also acetylene and coal-gas may 
 be used for the same service in welding aluminium. 
 
 Before welding, articles of aluminium usually have to be heated 
 up in the furnace to about 300° C, being covered with asbestos in the 
 furnace. As soon as this temperature has been reached, the article 
 should be drawn from the furnace, welded immediately, and, when 
 completed, returned to the furnace to be reheated to 300° C, then 
 allowed to cool till the next morning in the furnace, and kept free 
 from air to prevent shrinkage, cracks, and fractures. Many alumin- 
 
80 
 
 76 
 
 70 
 
 15 
 
 20 
 
 26 
 
 2 
 
 3 
 
 4 
 
 174 MODERN METHODS OF WELDING 
 
 ium -alloy castings may be welded without preheating, such as lugs 
 or projecting pieces broken off completely. There is a great variety 
 of alloy mixtures, many of which are found in welding shops in 
 articles such as gear cases, engine cases, chain cases from automobiles. 
 It is hard to judge accurately the composition of these alloys. 
 
 It is well to stock welding-rods of three different compositions, 
 which should be as near as can be to the same analysis as the articles 
 to be welded. From tests the author has made, the three following 
 mixtures can be used, and will give successful results if the articles 
 are graded to suit them : 
 
 Aluminium 
 
 Zinc 
 
 Copper 
 
 In welding aluminium-alloy articles, the rod must not be pure 
 aluminium. It must be of the same materials as the article to be 
 welded. When welding, a flux must be used the same as for welding 
 pure aluminium. Aluminium alloy is not a ductile metal. Hence 
 it must be treated as one treats cast iron, and the j)henomenon of 
 expansion and contraction has to be dealt with. 
 
 Fig. 87 is an alloy gear case, which plainly tells what parts are 
 broken and have to be made good. There are three cracks between 
 the cylinder openings. 
 
 The first procedure on such a casting is to clean it thoroughly and 
 free it from oil. Secondly, bevel the edges. Thirdly, prepare a 
 piece of sheet iron and fix inside the case under the cracks. This 
 must be larger than the cracks, to prevent the molten metal from 
 falling through. This will enable you, too, to penetrate right through 
 the weld. Having got this all prepared and placed on the welding 
 table (whilst cold), and put just in the position where welding will 
 take place, get the blowpipe, welding-rod, and flux all ready. Test 
 the hydraulic safety valve. See that you have enough oxygen and 
 acetylene. All being ready, and the tools at hand, place the article 
 to be welded in the furnace, and let it remain there till a temperature 
 of 350° C. is reached. Then remove from the furnace and place in 
 the exact position as when cold, and immediately commence to 
 weld. Then proceed regularly and progressively until the whole line 
 of welding has been done, adding at the same time, as the progression 
 takes place, equal amounts of the welding-rod to fill up level to the 
 top of the edges, dipping the hot rod in the flux from time to time. 
 The welding must be done quickly, and must not be gone over a 
 second time. The tip of the flame must not be allowed to touch the 
 
ALUMINIUM 
 
 175 
 
 metal. Immediately the weld has been completed, it should at 
 once be placed in the furnace again, and the temperature raised to 
 325° C. Afterwards it must be allowed to cool slowly, free from 
 any draughts or air. 
 
 After cooling, the article should be examined to see if the weld 
 has been homogeneous, and searched for any further cracks. There 
 should not be any if the heating has been uniform. The weld should 
 be cleaned up, or machined if required, and tested to see if it has 
 been distorted in any way. If not, the weld is satisfactory. 
 
 If operators will follow out the instructions above, they will 
 
 Fig. 87. — Fractured Aluminium-Alloy Gear Case. 
 
 succeed on every occasion. Each point must be watched, studied, 
 practised, and practised time after time. 
 The following are important : 
 
 (1) Blowpipe must not be too powerful; oxygen at its minimum 
 pressure; acetylene always slightly in excess. 
 
 (2) Articles must be well prepared, and placed in the welding 
 position before placing in the furnace. 
 
 (3) Articles must be heated to a temperature of 350° C. before 
 starting welding. 
 
 (4) Articles must not be allowed to go below the temperature 
 of 275° C. while welding. If they do, they must be put back in 
 the furnace and reheated to 350° C. The welding may then be 
 completed. 
 
 (5) After welding, the article must be put into the annealing 
 furnace, reheated to 350° C, and then allowed to cool slowly. 
 
176 MODERN METHODS OF WELDING 
 
 (6) Welding-rods must be approximately of the analysis of the 
 welded article, and must be free from impurities. 
 
 The chief uses to which magnesium is put are, as an alloy with 
 other metals, and for intense illuminations of short duration. When 
 alloyed with aluminium containing one or more other metals, the 
 crystallisation and other properties are modified. As a scavenging 
 alloy, it clears up oxide of other alloys. Because of its intense avidity 
 for both oxygen and nitrogen it is valuable in aluminium, nickel, 
 
 
 Fig. 88. — Aluminium- Alloy Gear Case Repaired. 
 
 copper, brass, etc., and in special steels. In aluminium castings, 
 for instance, less than 2 per cent, of magnesium cleans the metal, and 
 leaves from | to 1| per cent, in the casting, about doubling its tensile 
 strength, quadrupling its resistance to shock and jar, and producing 
 a much more easily machined metal. 
 
 Fig. 88 is an aluminium gear case, which had one corner of the 
 case broken, afterwards welded successfully. 
 
CHAPTER XXVIII 
 COPPER 
 
 Copper stands alone among the metals in having a reddish colour. 
 It is capable of taking a high polish, but on exposure to the air the 
 surface darkens considerably. It is comparatively soft (H=3), is 
 easily scratched with a knife, flows readily under pressure, is both 
 malleable and ductile, and can be rolled into thin sheets or drawn 
 into fine wire, and readily worked into any form by stamping and 
 spinning. 
 
 It is malleable, both cold and at a red heat, but near the melting- 
 point it becomes brittle. The tensile strength of cast copper is about 
 13 tons per square inch, but rods may be obtained having a strength 
 up to 26 tons, as mechanical working, especially wire drawing, greatly 
 increases its strength. When copper is worked, it becomes hard, 
 and loses its ductility to some extent. This can, however, be restored 
 by annealing. The specific gravity of copper is from 8-8 to 9 (various 
 figures are given by different authorities), depending on its state. 
 Castings have a lower specific gravity than sheets, and the specific 
 gravity of the latter is lower than that of wire. 
 
 Copper melts at 1,083° C. and boils at 2,310° C. This is im- 
 portant to remember as, if the heat in welding much exceeds the 
 melting-point, the copper will be burnt and full of blowholes. 
 
 It cannot be distilled. Its latent heat of fusion is about 44 
 and its specific heat roughly 0-094 ; but this increases as the tempera- 
 ture rises. The mean specific heat between 0° and 1° may be taken 
 as 0-0939 to 0-00001778; the heat required to raise 1 gramme from 
 0° to the melting-point and melt it would be 44+("094 x 1,085) = 
 146 units. The coefficient of linear expansion is 0-00001596 for 
 each degree centigrade rise of temperature. 
 
 Copper is an excellent conductor of both heat and electricity. 
 Its heat conductivity is 898, and its electric conductivity slightly less 
 than that of silver ; taking the resistance of the latter as 1, that of 
 annealed copper is about 1 -003, and that of hard drawn copper about 
 1-086. It is necessary to say " about," because the electric con- 
 ductivity is diminished very considerably by the slightest traces of 
 
 177 12 
 
178 MODERN METHODS OF WELDING 
 
 impurity. Copper can now be obtained so pure that the conducti- 
 vity is considerably greater than that taken for pure copper when the 
 standards in use were fixed. The resistance of a foot of pure copper 
 wire, 0-001 inch in diameter, is 9-612 ohms. The conducting power, 
 as in the case of all metals, falls as the temperature rises, the fall of 
 conducting power being 29-3 per cent, for a rise of temperature 
 from 0° to 100° C. 
 
 The principal varieties of commercial copper are : ( 1 ) electrolytic ; 
 (2) best selected (B.S.) tough. 
 
 Electrolytic copper is prepared by electro-deposition from solu- 
 tion and is usually very pure. It comes into the market precipitated 
 in cakes £ inch thick, deposited on both sides of a thin plate of 
 copper; or, after remelting, in ingots. Best selected copper is 
 mainly used for the manufacture of alloys, as it is now prepared 
 from pure materials. It is generally specified to contain not more 
 than 0-05 per cent, arsenic, a trace of antimony, and no other dele- 
 terious material. Tough copper is the name given in this country 
 to refined copper cast into slabs or billets for rolling into sheets, rods, 
 or tubes. It usually contains from 0-25 to 0-5 per cent, arsenic, 
 from 99-5 to 99-2 per cent, of copper, and only small quantities 
 of other impurities. 
 
 The British standard specification for the testing of copper is as 
 follows : 
 
 Copper Plates for Locomotive Fire-Boxes. 
 
 Tensile Mechanical Test. — A standard test-piece having a gauge 
 length of 8 inches must show a tensile breaking strength of not less 
 than 14 tons per square inch, with an elongation of not less than 
 35 per cent. 
 
 Bend Test. — Pieces of the plate shall be tested both cold and at a 
 red heat by being doubled over themselves (that is, bent through 
 an angle of 180°) without showing either crack or flaw on the 
 outside of the bend. 
 
 Stay-Bolts. 
 
 The rods must be clean, smooth, uniform in diameter, and 
 free from surface defects. The tensile test must not be less than 
 41 tons, and elongation not less than 40 per cent. In the hammering 
 or crushing down test, a piece of rod 1 inch long shall be placed on 
 end, and hammered and crushed down to a thickness of -| inch 
 without showing either crack or flaw on the circumference of the 
 resulting disc. 
 
COPPER 179 
 
 Copper Locomotive Tubes. 
 
 Tubes must contain not less than 99 per cent, of copper and 
 0-35 to 0-55 per cent, must consist of arsenic. Tubes must stand 
 bulging or drifting without showing either crack or flaw, until the 
 diameter of the bulged end measures not less than 25 per cent, 
 greater than the original diameter of the original tube. 
 
 Flanging Test. — The tubes must stand flanging without showing 
 either cracks or flaws until the diameter of the flange is not less than 
 40 per cent, greater than the original diameter of the tube. 
 
 Flattening and Doubling-Over Test. — The tubes must be caj^able 
 of standing both cold and a red heat, without showing either crack 
 or flaw. A piece of tube is flattened down until the interior of the 
 two surfaces of the tube meet. It is then bent so as to be doubled 
 
 Fig. 89. — Photograph or Section Across a Weld Performed without a 
 Special Phosphor-Copper Welding-Rod. 
 
 Notice the numerous blowholes. 
 
 over itself, bent through an angle of 180°, the bend being at 
 right angles to the direction of the length of the tube. 
 
 Hydraulic Test. — All boiler tubes shall be tested by an internal 
 hydraulic pressure of at least 750 pounds per square inch. 
 
 The oxy-acetylene welding of copper is not a stupendous job, 
 but is as easy as with any ordinary mild steel stocks, provided that 
 the necessary instructions are carried out in the operation of welding. 
 It is not possible to weld copper with an ordinary copper rod. The 
 copper when melted fuses, oxidation takes place, and the metal is 
 burnt through overheating, its temperature reaching boiling-point 
 in the attempt to make it weld. This leaves the copper weld full of 
 blowholes and badly oxidised, as the above illustration proves. 
 
 No matter what efforts are made to get good welds of copper, 
 with only copper rods they would not be a success. It is necessary 
 to have some flux to break down the oxide. The best method is to 
 
180 MODERN METHODS OF WELDING 
 
 incorporate the flux in the welding-rod, which will afterwards be 
 diffused in the molten mass as the melting takes place. If the 
 welding is done with a proper anti-oxidising rod, it will be quite 
 up to the other part of the article. 
 
 The welding-rod of copper should contain a very small per- 
 centage of phosphorus, with a trace of aluminium. The phosphorus 
 is mixed with the copper when the rods are manufactured, in small 
 quantities, evenly distributed throughout the rods, thereby securing 
 equal mixture in the line of welding. It is very important that the 
 proportion of phosphorus shall not be excessive, as this causes the 
 metal to lack fluidity, and also leads to loss of elongation. The 
 phosphuretted welding-rod is made in all sizes, from ^ to J inch 
 diameter (the latter is used for welding repairs in locomotive fire- 
 boxes). It is usually made in large, short, round bars, and drawn 
 to the various sizes, which are then cut off to a length of 24 inches 
 and bundled. 
 
 In the welding of copper articles it is usual to employ, in combina- 
 tion with the welding, a flux, or cleaning agent. There are several 
 compositions of these fluxes. One very good one, which is largely 
 used, consists of chloride of sodium 20 per cent., boracic acid 45 per 
 cent., sodium borate 35 per cent. Another for copper alloy is: iron 
 peroxide 35 parts, manganese peroxide 1 part, magnesium carbonate 
 \ part, alum 18 parts, silica 3 J parts, borax 4 parts. Mix and stir 
 well. Another consists of zinc oxide and charcoal in equal parts 
 mixed with molasses water to a stiff paste, formed into balls and 
 then dried. 
 
 Copper welds should be prepared just in the same manner as for 
 iron and steel; much more care and attention must be taken in 
 cleaning the edges to be welded. If they are not well cleaned, the 
 oxide or scale makes welding more difficult, sometimes causes adhe- 
 sion, or gets internally into the weld and causes blowholes. 
 
 With thin copper sheet it is imperative to have it supported 
 underneath. If this is not done, the metal soon runs through 
 owing to its fluidity, and it is very difficult to stop up the hole 
 that has been made. It is the custom in some workshops to use 
 a thick copper plate, which, on repetition work, assists the heating 
 of the article welded. Sometimes asbestos board is used, but the 
 author's experience is that asbestos wants renewing too often, as 
 the workmen seem to pull it to pieces quickly. 
 
 Further, there is a good smooth joint underneath, and welding- 
 may go right through. But in the case of asbestos, if it goes 
 through, the blowpipe usually burns or fires the asbestos, causing 
 
COPPER 181 
 
 a dazzling light, and often leaves rough holes or surface on the 
 underside. 
 
 The power of the blowpipe for copper welding should be one 
 size higher than that used for iron and steel, but the pressure of 
 the oxygen should be reduced to its minimum. The larger pipe 
 is needed because copper is a very high conductor. To counteract, 
 to some degree, this conductivity, it is very necessary, too, to heat 
 up the copper article before welding. 
 
 The diameter of the welding-rod should be according to the thick- 
 ness of the article to be welded, but slightly thicker than the same 
 thickness for iron and steel. The minimum diameter is 16-gauge, 
 but for general welding a stock of each size should be kept. A very 
 good flux for copper is 2 parts cryolite, 1 part phosphoric acid. 
 One has to be more careful in the welding of copper articles than 
 with iron and steel, although the same procedure has to be followed 
 out. An important point is the rapidity with which it must be 
 welded. 
 
 Also the weld must not be gone over twice, otherwise it will be 
 burnt, oxidised, and full of blowholes. Another point, which must 
 be watched, is that the melting-point of copper is 1,083° C, and the 
 boiling-point 2,310° C. This difference in temperature is vital. 
 When the melting-point is reached, welding must be proceeded with 
 quickly, and the blowpipe must be passed smartly over the line of 
 welding, so as to just melt and no more. The molten copper will be 
 in the form of a viscous, thick liquid. The addition of phosphuretted 
 rod will make a good clean weld, with a finish as smooth as the copper 
 article itself. There will be no oxide, no blowholes, and the weld 
 should stand any ordinary tests. 
 
 On the other hand, if the welding is done slowly, and the blowpipe 
 is held on the metal too long, the weld becomes too liquid by the 
 extra heating of the metal. When the temperature is raised to a 
 point near 2,310° 0., at which the copper boils, the metal is burnt. 
 Oxides form, which are absorbed in the metal as it cools, and cause 
 blowholes to form throughout the weld, making it defective and 
 useless. 
 
 Copper is the greatest conductor of heat of all metals. In 
 welding, preheat it in a preheating furnace, if the facilities are 
 at hand. Otherwise welding cannot proceed immediately owing to 
 the necessity of preheating with the blowpipe to make up for the 
 heat dispersing through the mass. 
 
 Procedurs in Copper Welding. — When the article has been pre- 
 pared, the blowpipe applied, and the welding-rod heldin the left hand, 
 
182 MODERN METHODS OF. WELDING 
 
 melting starts at one end (at the same time the welding-rod being 
 near the flame) ; the blowpipe is raised slightly to impinge the flame 
 on the rod, which melts into the molten weld and unites therein. 
 The blowpipe still continues melting in a progressive and continuous 
 manner. Never let the white cone of the flame touch the metal, 
 and take care to melt, not burn, the two edges of the weld. See 
 that the metal remains a viscous liquid, and does not become 
 " skilly," or thin liquid. Go through to the end of the weld 
 without increasing the melting-point. A good weld may be 
 hammered, bent, and annealed, and will stand the tests appearing 
 in the first part of this chapter. 
 
 The tip of the white jet of the flame should be kept y\ inch 
 from the metal. The treatment after welding is important, for 
 
 Fig. 90. — Microphotographs from the Region op a Weld Executed without 
 Special Welding-Rod. 
 
 Notice the separation of the crystals of oxide (deep black). The grey streaks in 
 the section on the right are the eutectic alloy of copper, containing 4 per cent, 
 of the oxide. 
 
 without it the copper is inclined to be brittle. The whole article 
 should be heated, and the weld vigorously hammered. After 
 hammering, heat the article again to a dull red heat and plunge 
 suddenly in cold water. 
 
 Heated copper combines with oxygen, forming what is known 
 as cuprous oxide. This oxide is absorbed in the molten copper, 
 under the normal action of the blowpipe, and it crystallises on cooling. 
 When oxidised to this extent, the weld is extremely fragile. The 
 only means of overcoming this is, as before described, the use of 
 a proper, skilfully prepared alloy welding-rod containing a very 
 small percentage of phosphorus. 
 
 The above illustration shows the oxide. 
 
 The phosphorus welding-rod seems to have the property of pre- 
 
COPPER 1S3 
 
 venting the formation of blowholes, either by suppressing the solu- 
 tion of the gases, or by aiding their evolution before the tempera- 
 ture of solidification. The phosphorus has the additional advantage 
 of giving rise to a protective varnish on the surface of the molten 
 copper. This is due to the formation of the oxide of phosphorus, 
 which combines with the oxide of copper, forming a fusible green 
 copper phosphate. This substance rises to the surface and protects 
 the copper from further action of the gases of the blowpipe flame. 
 Copper welds, when properly done, should be hammered where 
 possible at a red heat, and then reheated and plunged in cold water. 
 The sizes of copper phosphuretted welding-rods should be thicker 
 than for mild steel ; for instance, I inch copper should have a -f\ inch 
 diameter rod. As regards expansion, copper must be treated the 
 same as cast iron — that is, all articles must be preheated, and 
 annealed afterwards. 
 
 Failures in welding copper can only be caused by : 
 
 (1) Using a welding-rod of bad quality. 
 
 (2) The absence of flux when the welds are not absolutely clean 
 
 (3) Execution of the weld before the copper article is raised to 
 a high tenrperature. 
 
 (4) Bad joining of the metal and irregular feeding of the 
 welding-rod. 
 
 (5) The effects of expansion being badly opposed both during and 
 after welding. 
 
 (6) Getting the copper to too high a temperature, greatly 
 exceeding the melting-point. 
 
 (7) Going over the weld twice, not adding further welding-rod, 
 thereby causing oxidation and blowholes. 
 
 There are numerous applications in which the welding process 
 may be adopted. As regards loco fire-boxes, it has been successful 
 in some cases but not in others. Eor copper tanks, bends, rods in 
 electrical work, copper bars and rings on armatures it is being found 
 useful. Household copper boilers are being extensively welded; 
 and there are many other directions in which welding could now 
 take the place of brazing. 
 
CHAPTER XXIX 
 BRONZE 
 
 Bronze may be considered as an alloy of copper and tin, the former 
 element predominating. Alloys with 1 to 2 per cent, of tin show 
 nearly the ductility of pure copper. They can be worked in the 
 cold state under the hammer more readily than pure copper. The 
 ductility decreases rapidly with an increase in the content of tin. 
 An alloy containing 4 per cent, can only be worked with the hammer 
 at a red heat, and soon cracks when one attempts to hammer it 
 cold. Alloys containing up to about 15 per cent, of tin can no longer 
 be hammered, even in a warm state. Alloys with about 9 per cent, 
 of tin show the greatest strength of all bronzes, and those with 
 about 15 per cent, possess the greatest hardness and strength. The 
 maximum of hardness and brittleness lies between 28 and 35 per 
 cent, of tin. There are various bronzes on the market, those having 
 a percentage of aluminium, manganese, phosphor, etc., being known 
 by a double name — aluminium bronze, manganese bronze, etc., 
 respectively. Some of these alloys are true bronzes, as they often 
 contain no tin. 
 
 Aluminium Bronzes. 
 
 The proportion of aluminium alloyed with copper varies from 
 1 to 10 per cent. The alloys are as strong as mild steel, highly 
 malleable, elastic, and ductile. The presence of other metals 
 impairs the quality. An alloy containing 10 per cent, has a tensile 
 strength of 40 to 45 tons per square inch. 
 
 Manganese Bronzes. 
 
 These contain copper, manganese, zinc, and tin, and sometimes 
 they are characterised by hardness, elasticity, and strength, com- 
 bined with toughness and resistance to corrosion. They can be 
 rolled and forged hot. An important application is for the propellers 
 of steamships. They are also used in general engineering brass- 
 work. The manganese is generally introduced in the form of ferro- 
 manganese or as manganese copper. 
 
BBONZE 185 
 
 Phosphor-Bronze . 
 
 Phosphor-bronze contains a small proportion of phosphorus, 
 introduced either as a phosphor-tin (obtained by dissolving 
 phosphorus in molten tin up to 20 per cent, of phosphorus) 
 or as phosphor-copper, after fusion, or the ordinary ingredients. 
 The tin varies from 4 to 10 per cent, and the phosphorus from -1 to 1 . 
 Where toughness and ductility are required the phosphorus should 
 not exceed 0-1. Metals containing more, increase in hardness and 
 are used for valves, bushes, cogwheels, etc. Phosphorus should 
 be cast at as low a temperature as possible. 
 
 Silicon Bronze. 
 
 Silicon bronze contains silicon and is harder and stronger than 
 ordinary bronze. The beneficial effects of phosphorus and silicon 
 are generally attributed to the powerful deoxidising influence they 
 exert on account of their affinity for oxygen. Bronzes do not absorb 
 the heat like copper, although they absorb it more than iron and 
 steel. There are several mixtures of bronzes, and one must be care- 
 ful to use a welding-rod which is nearly the same mixture as the 
 bronze article being welded. The table on p. 162 shows several 
 varieties. 
 
 Welding- Rods. — Welding -rods destined for the efficient welding 
 of bronzes must be carefully manufactured from very pure metal, 
 and their constituents must be the same as the article to be welded, 
 with a small addition of phosphorus and a trace of aluminium. 
 These rods should be carefully mixed when they are made, must not 
 contain any impurities whatever, and should be made from new 
 materials. They are made in sizes from jr to \ inch diameter and 
 24 inches long. They should be sand-blasted if cast, so as to free 
 the rods from the gritty sand. 
 
 In the welding of bronzes it is necessary to have a cleaning flux 
 to scour the metal as it becomes molten. The flux can be obtained 
 from chemists who are skilled in the compounding of these mixtures. 
 A good flux for bronze is equal parts of phosphoric acid and 80 per 
 cent, alcohol. Others are equal parts of crude tartar and nitre 
 burned together; and 3 parts nitre, 2 parts argol. 
 
 Bronze Welding. — The first operation when one has a bronze 
 article is to see if it is bevelled. If not, this must be done, and the 
 weld properly cleaned and freed from all grease. It is usual to 
 place the articles in the preheating furnace to get them hot to 
 assist welding, and to prevent fracture from uneven heating. It is 
 
180 MODERN METHODS OF WELDING 
 
 important if the article is unsupported on the inside to support it, 
 as unless this is done the weld cannot be penetrated right through. 
 The power of the blowpipe must be one size higher than that for iron 
 and steel of the same thickness. The blowpipe must be properly 
 regulated, and the oxygen must not be in excess, otherwise the weld 
 would be oxidised and burnt. Likewise, if the acetylene is in excess, 
 carbonisation will occur. The blowpipe must be well regulated until 
 a clear white cone is reached, neither oxidising nor carbonising. 
 Attention must also be paid to the pressure of the oxygen, which 
 must not be more than that stated by the makers of the blowpipe. 
 This is very important. 
 
 Method of Welding. — The article should be fixed up usually in a 
 horizontal position. If broken, the parts must be secured to pre- 
 vent them from being out of line when welded. Proceed with the 
 blowpipe (already properly regulated) to heat the edge of the line 
 of welding, about \ inch from the actual edge, and start melting 
 the two bevelled edges at this point. As soon as they become 
 molten, add a little welding-rod, to which is annexed the flux, and 
 then bring the blowpipe to the edge of the weld. Bring this to a 
 molten state, add welding-rod, fill up bevel, and proceed forward 
 with the welding, at the same time giving a regular gyratory move- 
 ment. Both edges of the weld must be melted together simul- 
 taneously with the welding-rod with an occasional dip in the flux. 
 See that the rod is kept sufficiently in the molten metal to fill up 
 the bevelled edges to the same thickness as the article being welded. 
 One must not go over the weld twice without adding fresh metal. 
 If this is done oxidation takes place and the weld is full of blow- 
 holes and spoilt. 
 
 As soon as the weld is done, the article must be put into the 
 annealing furnace and heated up to about 600° C. and then allowed 
 to cool down slowly. 
 
 Failures in welding bronzes are due to : 
 
 (1) The use of an impure welding-rod. 
 
 (2) The overheating of the metal, making it too fluid and caus- 
 ing oxidation and blowholes. 
 
 (3) The weld being gone over twice, leading to oxidation. 
 
 (4) Insufficient penetration, causing adhesion. 
 
 These failures can easily be overcome if operators practise regu- 
 larly, and test their pieces until they find out that they have become 
 proficient. 
 
CHAPTER XXX 
 
 BRASS 
 
 Brass is an alloy of copper and zinc, and is in most general 
 use. It should only contain copper and zinc, but most varieties 
 contain small quantities of impurities. Copper and zinc can be 
 mixed together -within very wide limits, the resulting alloys being 
 always serviceable. 
 
 Generally speaking, it may be said that with an increase in the 
 content of copper the colour inclines more to golden, the malle- 
 ability and softness of the alloy being increased at the same time. 
 With an increase in the content of the zinc the colour becomes lighter 
 and finally shades into a greyish-white, while the alloy becomes 
 more fusible, more brittle, and at the same time harder. The physi- 
 cal properties of brass varj^ according to the relative quantities of 
 copper and zinc. Alloys containing up to 35 per cent, of zinc can be 
 converted into wire and sheet in the cold state only, those with from 
 15 to 20 per cent, being most ductile. Alloys with from 36 to 40 
 per cent, of zinc can be worked in the cold state as well as the hot. 
 With a still higher content the ductility increases rapidly; and an 
 alloy, for instance, from 60 to 70 per cent, of zinc is so brittle that 
 it cannot be worked. If, how r ever, the zinc is increased up to a 
 maximum (70 to 80 per cent.), the ductility increases again, 
 and the alloy can be worked quite well in the hot state, but 
 not at red heat. The strength of brass is intimately connected 
 with its composition, that containing about 28-5 per cent, of 
 zinc showing the greatest absolute strength. The strength depends 
 to a great extent upon the mechanical treatment the metal has 
 received. 
 
 An important factor in brass is its melting-point, there being 
 wide deviations in this respect, which are readily explained by the 
 great difference in the melting-points of the two constituent metals. 
 Generally speaking, the fusing-point of brass lies at about 1,832° F. 
 The mixtures for certain purposes are legion. 
 
 Hot-rolling 70/30 Brass.— It is quite possible to roll this metal 
 
 187 
 
188 
 
 MODERN METHODS OF WELDING 
 
 hot by using electrolytic copper and an electrolytic spelter. 
 
 The following is the mixture : 
 
 Pounds. 
 
 .. 68 
 
 2 
 
 . . 30 
 
 Electrolytic copper 
 Cupro-manganese (25 per cent.) 
 Brunner-Mond electrolytic spelter 
 
 The metal rolls well at a bright red heat, but is more expensive 
 than ordinary brass. 
 
 Admiralty Specification for Brass. 
 
 The composition of the mixtures used throughout the whole of 
 the work supplied is to be as follows. New metal only is to be used 
 for castings subject to steam pressure. 
 
 Tin. 
 
 Zinc. 
 
 Copper. 
 
 Naval brass 
 
 Brass for condenser tubes 
 
 37 
 29 
 
 62 
 70 
 
 All brass subject to the action of sea water must not in any case 
 contain less than 1 per cent, of tin. 
 
 Tensile test-pieces taken from castings must stand the following 
 
 tests : 
 
 Ultimate Tensile 
 
 Strength Not Less 
 
 Than 
 
 Naval brass round bars, £ 
 Naval brass round bars, above f 
 High-tension brass, forged 
 High-tension brass, cast 
 
 14 tons 
 26 ,, 
 22 ,, 
 
 28 ,, 
 
 Elongation in 
 Length of 2 Inches. 
 
 1\ P er cent. 
 30 
 30 
 25 
 
 Naval bars must be capable of : 
 
 (a) Being hammered hot to a fine point. 
 
 (b) Being bent cold through an angle of 75° over a radius equal 
 to the diameter of the bars. Breaks within less than | inch of the 
 grip are not accepted. 
 
 General. — The metal castings are to be sound, clean, and free 
 from blowholes, and no piecing, parching, or stopping will be per- 
 mitted. 
 
 The above articles are often welded, and one must study their 
 elements in order to be able to execute satisfactory welds on them. 
 
BRASS 
 
 189 
 
 Brasses are bad conductors, but quite as fluid as copper. There are 
 two classes of brass — first and second qualities. The first melts at 
 930° C.j the second at 880° C. This is important, as if they are got 
 to high temperature they burn, the zinc volatilises, and the metal 
 is oxidised. The melting of brass under the action of the blowpipe 
 is accompanied by the phenomena of the absorj)tion of gases, 
 volatilisation of zinc, and oxidation. 
 
 Operators must not attempt to weld brass autogenously with- 
 out understanding the metallurgical idiosyncrasy, otherwise success 
 will not be obtained. Much trouble springs from the volatilisation 
 of part of the zinc, which rises in dense white fumes. Blowholes 
 then abound, part of the zinc is lost, and the weld has no strength. 
 
 Welding-rods for the welding of brasses are manufactured in the 
 same way as copper rods, and should be as near as possible in com- 
 position to the metal being welded, so as to keep the same mixture 
 of alloy as in the article itself. It is usual in shops where a variety 
 of brass castings are welded to keep sets of welding-rods made up 
 to the different analyses to suit the castings. These rods can be 
 obtained from the various manufacturers who make and stock them. 
 
 Rods for brasses are made from the purest new metal, in which 
 a minute quantity of aluminium should be evenly incorporated 
 throughout the length of the rod. They are made in various sizes, 
 from ^ to | inch thick. In conjunction with the rod a cleaning flux 
 must be provided, which dissolves the alumina, prevents the vola- 
 tilisation of the zinc, and protects the metal from oxidation. There 
 are many fluxes in use, but for general purposes the following give 
 good service: 
 
 [ Chloride of sodium .. .. 30 per cent. 
 
 (1) -Boric acid 
 (Borax 
 
 ii\\ | Selenium 
 (wj \Charcoal 
 
 [Sodium carbonate 
 /q\ | Silica (white sand) 
 ^ ' 1 Coal dust (anthracite) 
 
 [Bone ash 
 
 40 
 
 55 
 
 30 
 
 55 
 
 2 
 
 parts 
 
 1 
 
 part. 
 
 5 
 
 parts 
 
 15 
 
 55 
 
 5 
 
 55 
 
 20 
 
 55 
 
 Execution of Brass Welds. 
 
 The edges are prepared as for iron and steel, according to the 
 thickness, on one side or both. If it is a case of castings, these two 
 must be bevelled wherever the crack or break occurs, otherwise one 
 is sure to get adhesion or lack of penetration, excessive consumption 
 
190 MODERN METHODS OP WELDING 
 
 of gases, burnt metal, and a defective weld. This bevelling cannot 
 be insisted on too much. The bevel must be 90° to the thickness 
 of the metal — that is, 45° each side. Brass, more than metal 
 (aluminium excepted), must be thoroughly cleaned, otherwise 
 adhesion takes place. Brass welds are easy to do and give good 
 all-round results, but only when the weld line is properly bevelled 
 and clean. Also it is advisable to preheat the article to just under 
 500° C. 
 
 The blowpipe used is the same as in the case of copper: one 
 size larger than for iron and steel. The size has been selected to 
 suit the thickness of the articles being welded. For instance, 
 75- inch thick requires a blowpipe consuming IT cubic feet of acety- 
 lene per hour — that is, size 2 blowpipe. The welding-rod for ^-inch 
 thick metal should be ^ inch diameter. 
 
 It is essential to have a rod a little thicker than the metal to be 
 welded in the case of brass. Many brass articles require to be 
 supported underneath the welding line, to allow good penetration of 
 the weld. In welding, the blowpipe should be started a little from 
 the end (about | inch), so as to preheat this part prior to doing 
 the starting edge. When this part is beginning to become molten, 
 just put your blowpipe on to the edge. At the same time add the 
 welding-rod to fill and cover up the bevel, and then proceed along 
 the line of welding, melting both edges of the article and the feeding- 
 rod at the same time. Continue advancing with a gyratory move- 
 ment and adding with regularity the welding-rod, which must be 
 dipped in the flux from time to time, according to progress made. 
 Welding must be continuous until the whole line of welding is com- 
 plete. The welding must be rapid, far faster than with iron or 
 steel. Care must be taken not to let the blowpipe stay too long- 
 on the weld, or it will burn, and will increase the temperature, and 
 cause the metal to boil and bubble, which no flux or rod will check. 
 Consequently the weld will be full of blowholes, the zinc will be 
 reduced, and the weld will be oxidised, and therefore useless. 
 
 To avoid these failures one must not use a rod of bad mixture, 
 nor let the blowpipe remain too long on the weld, otherwise the metal 
 " boils." The white jet must not touch the metal within ^ inch. 
 Brass of the first quality can be hammered and annealed, and can 
 stand energetic forging or stamping without showing fracture. 
 This hammering improves considerably its mechanical properties. 
 Brasses containing 55 to 65 per cent, of copper should be forged hot, 
 those containing 65 to 70 per cent, of copper hammered cold. After 
 hammering, brass should be annealed to just under 500° C. 
 
CHAPTER XXXI 
 AMERICAN METHODS 
 
 Operators will benefit by an insight into what the Americans are 
 doing in the way of welding processes. Shortly after the United 
 States entered the war, the Council of National Defence appointed 
 an Engineering Committee, which undertook as one of its chief 
 tasks a study of the possibilities of electric welding in shipbuilding. 
 This Committee was taken over by the Emergency Fleet Corpora- 
 tion. It later broadened its scope so as to include gas and thermit 
 welding. By its discussions, researches, lectures, and conferences, 
 its interchange with foreign countries and its dissemination of in- 
 formation, the Committee gave a great impetus to the use of welding 
 in America. 
 
 Concurrently with this work, another advance in welding, with 
 an emphasis on gas welding, was started by the formation of the 
 National Welding Council. With industry becoming normal, after 
 the Armistice, it was considered desirable to extend into the entire 
 field of joining metals, and an effective way of accomplishing this 
 was the foundation of the American Association for the Welding 
 Industry. 
 
 The Society was brought together in the manner usual with 
 scientific societies : persons from all branches of the industry interested 
 in any of the welding processes — forge welding, electric resistance 
 welding, thermit welding, gas welding, electric spot, butt, seam, and 
 arc welding. It will create and assist in maintaining the Bureau 
 of Welding, which will be a separate organisation, designed to take 
 full advantage of the principle of co-operation in research and 
 standardisation. Some of the Society's investigations may be cited. 
 
 (1) In arc welding a determination of the best current with 
 various sizes of electrodes. 
 
 ■ (2) A determination of proper methods of procedure in making 
 arc and gas welds, with a system of inspection as the work pro- 
 gresses. 
 
 (3) The acquirement of further knowledge of the characteristics 
 
 of metal as affected by welding, particularly its ductility and its 
 
 action under repeated tests. 
 
 191 
 
192 MODERN METHODS OF' WELDING 
 
 (4) In gas, spot, or arc welding methods of assembling large 
 structures to eliminate initial or locked-up stresses, due to contraction 
 on cooling. 
 
 (5) The ascertainment of a proper standard for both producer 
 and consumer. For some years numerous overlapping and con- 
 flicting efforts in the field of industrial standardisation have been 
 carried out by more or less independent organisations, each according 
 to its own methods of procedure. There was need, therefore, of 
 reducing unnecessary effort, and providing uniform methods that 
 will secure in each case the co-operation and support of all the 
 organisations whose interests may be affected. 
 
 Fig. 91. — Duograph Welding Machine. 
 
 Training of Operators. — The investigations of the Welding Com- 
 mittee have thus far shown that one of the most important elements 
 in the success of an autogenous welding operation is the skill of the 
 operator. To secure this, uniform methods of training are essential. 
 The Society is taking an active part in planning how operators should 
 be trained and how their proficiency may be determined. Some 
 of the objects of the American Society are as follows: 
 
 (1) To advance the art and science of welding. 
 
 (2) To afford its members opportunities for the interchange of 
 ideas with respect to the sciences and art of welding, and for the 
 publication thereof. 
 
 (3) To conduct researches into welding, co-operating with other 
 
AMERICAN METHODS 
 
 193 
 
 societies and associations, and with the Governmental departments, 
 for the benefit of the industry generally. 
 
 The illustration on previous page is of an automatic seam-weld- 
 ing machine, which is known as the "duograph." This machine, 
 specially designed for welding the seams of drums or containers, 
 ensuring a mechanical weld uniform in appearance and efficiency, 
 comprises a turret-top holding device, with water-cooled arms and 
 clamps for holding the steel drum in position, permitting of the 
 form being placed in position for welding on the one set of arms, while 
 
 Fig. 92. — Showing Different Operations of the Duograph. 
 
 the form on the opposite set is being welded. The turret top is 
 then swung half-round, the welded form removed, and another set 
 up ; this preparation requires less time than the actual welding of the 
 form in position. The blowpipe carriage is moved forward at a fixed 
 speed of welding, by a power-belt motor driven, and is reversed by a 
 hand wheel when the weld is finished. 
 
 Variable speed for different thicknesses of welding is obtained 
 by the use of cone pulleys. The blowpipe carriage is fitted with two 
 blowpipes, one above and the other below, for welding both sides 
 of the seam simultaneously. For very light welding, one blowpipe 
 
 13 
 
194 MODERN METHODS OF WELDING 
 
 only is required, welding from one side. Water-cooled blowpipes 
 are used, connected with rubber tubing to the water-supply. 
 
 This machine will weld a seam 36 inches long. An average speed 
 of welding is 18 inches per minute, or 90 feet per hour may be obtained 
 on 16-gauge sheets. 
 
 The photograph on p. 193 shows different operations of the duo- 
 graph. 
 
 Fig. 93 is of a medium-pressure acetylene generator, 15 pounds 
 pressure as a maximum. This is designed to provide acetylene 
 under suitable conditions and with proper control, to meet the 
 requirements of the oxy-acetylene welding and cutting, and to 
 make possible the employment of the positive-pressure principle, 
 utilising acetylene under direct and appreciable pressure, employ- 
 ing lump carbide for the generation of the gas, with indepen- 
 dent power for feeding the carbide, the feeding mechanism being 
 controlled by the gas pressure as generated : the carbide drops into 
 large voHmes of water, cools the gas, and generates slowly. 
 
 Safety has been given even greater consideration in the con- 
 struction of these plants than efficiency. They have automatic 
 feed, carbide to water, with independent power, and the quan- 
 tity of carbide remaining in the hopper is constantly indicated. 
 The acetylene gas is piped directly from the generator under 
 requisite pressure through a service pipe-line to the welding 
 stations, and regulated by reducing valves, fitted with pressure 
 gauges to govern the proper working pressure. 
 
 These are made in many sizes to most economically meet the 
 requirements of the purchaser, and may be had with capacity of 
 25, 50, 100, 200, and 300 pounds of carbide constituting a full charge. 
 These generators are exceptionally economical compared with the 
 low-pressure system. The latter are used almost exclusively in 
 England. The blowpipes used in low-pressure consume from 1-5 
 to 1'3 parts of oxygen to 1 part of acetylene. The medium-pressure 
 generators consume 1 part of oxygen to 1 part of acetylene. 
 Hence, therefore, medium -pressure generators save 40 to 50 per cent, 
 of oxygen over the low-pressure, also the medium-pressure gives 
 40 per cent, more welding. Their blowpipes never back-fire; they 
 keep in all day. Perfect welds are obtained ; no adhesion, no oxida- 
 tion, fully penetrated welds with 98 per cent, tensional stresses. 
 These advantages are attainable from the generators referred to, 
 which are known as the Davis-Bournonville. They are built from 
 heavy steel plates, and galvanised after manufacture. They are used 
 in very many workshops in the U.S.A., and give every satisfaction. 
 
AMERICAN METHODS 
 
 195 
 
 This medium-pressure avoids the defects of the low-pressure 
 types, which depend solely on the injector principle for the proper 
 mixture of the gases imperative to securing the best results. It will 
 be readily understood that in successful welding a neutral flame 
 must be employed, because if there is any excess of oxygen the metal 
 
 Fig. 93. — 200 Pounds Carbide Capacity, using If Pounds Carbide. 
 Height, 104 inches; diameter, 37 inches; weight, 850 pounds. 
 
 will be burnt and oxidised. By putting the gases under pressure 
 instead of depending largely on the injector principle, a positive 
 control of the gas is obtained. 
 
 Some of the most extensive work carried out in America was a 
 twelve months' job for eleven welders, welding a metal-roofed flue, 
 
196 MODERN METHODS OF WELDING 
 
 850 feet long and 120 feet wide, constructed with 690,000 No. 9 
 gauge plates. This job employed 200 tons of carbide in two 200- 
 pounds capacity generators by Davis-Bournonville Company, with 
 330,000 feet of oxygen and 9,200 pounds of y\-inch thick welding- 
 rods, for 53,365 lineal feet — or over ten miles of welding. 
 
 A notable welding job was the saving of 9,000 feet of 4 to 6 feet 
 diameter power pipeline, averting a loss of several hundred thousand 
 dollars, in the Colorado Mountains. The joints were leaking, and 
 over 4,000 feet of the heaviest portion of the pipe had been discarded 
 as impracticable; but the workmen carried on for several months 
 during the winter. 
 
 Nearly a mile of cutting was recently done by a single Davis- 
 Bournonville operator in Northern Canada. About two weeks' 
 intermittent operation was required, averaging about 350 lineal 
 
 Fig. 94. — Part Section, Non-Flash Blowpipe. 
 
 feet per day, making 5,097 lineal feet of cutting of a steel pipeline, 
 12 feet in diameter, 1,480 feet long. This was cut into two half- 
 sections, and the upper half into sections for removal. The last 
 85 feet was done in sixty-five minutes. 
 
 Range boilers are welded at 30 feet per hour. 
 
 Barrel seams (30 inches long, by 20-gauge thick) are welded at 
 the rate of 75 to 85 seams per day. 
 
 An immense hydraulic press was repaired at a cost of about 
 $350, saving the owners about $1,000 per day for sixteen 
 days. 
 
 A Corliss compound engine, 2,500 h.p., was repaired in four days. 
 There was one crack 14 inches long, another 23 inches long, and still 
 another 5 feet long. It was estimated that the cost of a new casting 
 would have been $2,000, not including the cost of dismantling and 
 reassembling; and the new casting could not have been procured in 
 less than six weeks. 
 
AMERICAN METHODS 
 
 197 
 
 Others that would heretofore have been considered marvellous 
 operations have been performed by the oxy-acetylene process, 
 but sufficient has been said to convince the readers of its 
 adaptability. 
 
 Another novel but useful tool is a blowpipe just brought out in 
 America which is known and sold as the " rego " bloAvpipe. It is 
 claimed that these torches do not, and cannot, flash bach. This is 
 owing to an arrangement in the blowpipe which is designed for 
 special supply and mixing chambers. The acetylene must be higher 
 
 Fig. 95. — Air-Gas Preheating Torch 
 Flame Playing on Mixing Chamber 
 oe Welding Torch until Tip and 
 Nut are Red-Hot. No Flash. 
 
 Fig. 96. — Welding Torch Tip Directly 
 on Metal at White Heat, Held 
 There till it Penetrates. No 
 Flash. 
 
 pressure than the oxygen, but not exceeding 15 pounds. The claim 
 is made that this invention imparts to the acetylene a sufficient 
 speed as it enters the mixing chamber and commingles with the 
 oxygen to ensure that at this point, with a neutral flame burning at 
 the tip, the speed of both gases shall be greater than that of the 
 flame propagation of the mixtures at the point where the gases 
 commingle. 
 
 The acetylene must be under greater pressure than that of the 
 oxygen to obtain this result. The heating of the chamber at this 
 point to a sufficient degree to ignite the mixed gases will not cause 
 a " flash back," since the speed of the mixture will carry the flame 
 
198 
 
 MODERN METHODS OF WELDING 
 
 to the tip. If obstructions, such as flying particles of molten metal, 
 or the bringing of the blowpipe close, or up to, the metal, reduce 
 the velocity of the mixtures at the tip and tend to drive the flame 
 into the interior of the pipe, the acetylene, being under greater pres- 
 sure, immediately seals the oxygen, causing a carbonising mixture 
 to flow from the mixing chamber to the tip and ignite. As the car- 
 bonising flame cannot flash back, the acetylene alone, or with some 
 quantity of oxygen (depending upon the size of the obstruction), 
 continues to burn at the tip until the obstruction is removed, when 
 the oxygen again flows through in full volume, producing a natural 
 flame. 
 
 The American Welding Committee of the Emergency Fleet Cor- 
 poration, under the chairmanship of Professor C. A. Adams, has been 
 
 Fig. 97. 
 
 -Welding Torch Tip Directly against Brick. No Escape for Heat 
 Waves except against Tip. No Flash. 
 
 of great assistance to the welding industry. At the second meeting 
 held, a communication was received from the U.S.A. Shipping Board, 
 requesting information and advice on the most economical method 
 of producing anchor chains in large quantities. A meeting of repre- 
 sentatives of chain manufacturers was arranged, and the work was 
 put in hand. In six weeks a sample was submitted. Within six 
 months production on an order of $1,000,000 for chains made by a 
 new process saved the Government $£0,000 at the start. This new 
 chain, made from cast steel, refined in an electric furnace, not only 
 met the specifications of the carefully hand-forged chains of the past ; 
 but in place of the average production by a gang of chain- welders of 
 the highest skill of less than 1,000 pounds per day, a foundry unit 
 with a 10-ton electric furnace produced 70 tons of 2-inch chain in 
 twenty-four hours, with mostly unskilled labour. 
 
AMERICAN METHODS 
 
 199 
 
 At the same meeting plans for spot welders of the portable type, 
 for from J up to 1 inch plates, were discussed. Under the leader 
 ship of H. M. Hobart, the Research Committee investigated the 
 current density suitable for various electrodes; non-destructible 
 methods of testing welds; the effects of locked-up stresses in weld- 
 ing long sections by rigid or non-rigid methods; the methods of 
 holding plates during welding ; the effect of corrosion on welds and 
 adjacent metal, conducting a series of tests on J-inch plates welded 
 by employees of the manufacturers 
 who supplied the apparatus — the 
 choice of electrodes, current density, 
 and method of control being left to 
 the discretion of the welders. They 
 finally submitted standard methods 
 for testing electrodes for welds of all 
 kinds, and revised specifications for 
 electrode wire. These tests are known 
 as the Wirt- Jones. As the result of 
 this instruction propaganda, the Ship- 
 yard Visiting Committee reported 
 that at Hogg Island alone hundreds 
 of thousands of parts were being 
 welded instead of riveted, the saving 
 being approximately 70 per cent. 
 
 Electric welding is used very ex- 
 
 tensively in America; the General ^^-^^t^Z^T^ 
 . J ' 5-Inch Axle, Tip in Hole. JNo 
 
 Electric Company are extending the Escape for Sparks or Heat 
 production of equipment for electric except against Tip. No Flash 
 welding for all classes of work. I 
 propose to show some of the equipment in general use later. 
 
 During the past few years the extension of welding of all kinds 
 to the building and repair of ships has been phenomenal, especially 
 as regards electric welding. While electric welding has been used 
 chiefly for iron and steel, the technique of the art for cast iron and 
 the various non-ferrous alloys employed in shipbuilding is being 
 rapidly developed. The welding of copper can be done with the 
 carbon electrode. Brass and bronze castings and flanges welded to 
 the pipes are very common. 
 
CHAPTER XXXII 
 THE METALLURGY OF ARC WELDING 
 
 We have learned to know, with a fair degree of certainty, what a 
 steel casting should be to be acceptable for any given engineering 
 purpose. We are apt to be very particular about casting when human 
 life would be in danger by its failure in service. This is true of all 
 iron and steel parts, which go largely to the make-up of our present- 
 day necessities. 
 
 In building up and bringing together many scattered facts about 
 the behaviour of iron and its alloys, under varying conditions, the 
 microscope has played a very important role. It satisfies the natural 
 curiosity to " see what's going on." Merely to see a line of signal 
 flags on a destroyer, however, does not help us much unless we know 
 what the signals mean. Just so an intelligent investigation of a 
 metallurgical product, like a weld made by an electric arc, involves 
 considerably more than a mere examination of the metal or its frac- 
 ture under the microscope, much as this may reveal. 
 
 The making of a good weld is essentially a metallurgical problem. 
 More specifically an arc weld is a steel casting made by a continuous 
 process both as regards melting and casting, i What we require is a 
 sound, fine-grained casting, free from blowholes and slag inclusions 
 and low in impurities. The casting must also make a continuous 
 and perfect union with the plate or material to be welded. The 
 physical properties of the weld will depend upon five distinct factors, 
 namely: (1) Crystal structure; (2) gas-holes; (3) slag inclusion; 
 (4) impurities ; and (5) composition. These factors are identical with 
 those determining the properties of any steel product, with the ex- 
 ception that most of the latter may be improved by heat treatment 
 or working, while in a large majority of cases the weld must be used 
 as made. The order in which these factors are given is not to be 
 taken as the order of their importance; the time has not arrived 
 when such an order can be set down. 
 
 Crystal Structure. — In studying the crystal structure of a large 
 number of welds, as revealed by fracture, it appears that a very 
 fine grain is produced by depositing the metal rapidly in compara- 
 
 200 
 
THE METALLURGY OF ARC WELDING 
 
 201 
 
 tively thin layers, thus preventing the plate from heating up suffi- 
 ciently to slow down the cooling. As soon as this occurs, columnar 
 crystals begin to form, with a resulting brittleness. 
 
 Fig. 99. — Showing Areas at High Magnification. 
 
 It is often desirable for other reasons, however, to maintain 
 as large a molten pool as possible. In such a case, the only way 
 to maintain a fine structure is to hammer the weld while hot, to 
 prevent the formation of too coarse a structure. The cooling effect 
 
 Fig. 100. — Nitride Areas in Electrolytic Iron treated as Fig. 102. 
 
 of the plate upon the weld structure may be readily observed in 
 running a short length of weld across a plate. The first part of the 
 weld will show a fine-grained fracture, while a little farther along 
 
202 
 
 MODERN METHODS OF WELDING 
 
 a decided growth begins, gradually changing to an entirely coarse 
 structure as the plate heats up. Methods of keeping the plates 
 cool with a stream of water have been tried with considerable success 
 
 Fig. 101. — Electrolytic Iron, Nitrogenised by Annealing in NH 2 Twelve 
 
 Hours, 750° C. 
 
 as far as the grain-size is concerned, but difficulties of manipulation 
 have prevented its adoj)tion in practice. 
 
 Gas-holes are to be found in all electric welds, and are an inipor- 
 
 a* 
 
 p' 1 
 
 
 •' * y 
 
 
 tot/'' ' - 
 
 & • ■ ■ 
 
 
 -■ y.-^.t ... 
 
 ■ ■■'■"''. ,' k*i 
 
 
 
 'V : vr^ . 
 
 '• '-' ' - *'\ 
 
 
 
 ■ : '* >-". : ■ 
 
 '... ^3 
 
 Fig. 102. — Electrolytic Iron, Nitrogenised by Annealing in NH 2 Twelve 
 Hours, 750° C, but subsequently Annealed in Vacuum for Two Hours 
 at 1,000° C. Areas Smaller and Lines Diminished in Number. 
 
 tant source of weakness. Their occurrence is frequently attributed 
 to the presence of dissolved gases or gas-forming impurities in the 
 electrode material. This is undoubtedly true, but only to a limited 
 
THE METALLURGY OP ARC WELDING 
 
 203 
 
 extent. Dissolved or occluded gases in electrodes are largely liber- 
 ated as the metal passes through the arc stream, and cannot have 
 any considerable effect upon the deposited metal. They do affect 
 
 Fig. 103. — Edge of Weld made with Covered Electrode, showing Coarsen- 
 ing op Grain in Plate Stock by Overheating. (Unannealed.) 
 
 the working of the electrode, however, as they cause spluttering, 
 frequently so bad as to make the electrode useless. Experiments 
 have shown this to be particularly true of highly oxygenous 
 
 Hit* I 
 
 m 
 
 j 
 
 :r*>* 
 
 
 H 
 
 m mm 
 
 mm? 
 
 Eig. 104. — Unannealed Weld Section. Weld made with Standard Bare 
 
 Wire Electrode. 
 
 electrode steel. Carbon is one of the worst offenders in producing 
 gas-holes. Always ready to combine with oxygen, it finds a rich 
 supply in the metal deposited by the arc. Carbon monoxide is 
 
204 
 
 MODERN METHODS OF WELDING 
 
 formed and, owing to the rapid solidification of the metal, is trapped. 
 The carbon in the plate also becomes an important factor in this 
 connection. The carbon in that portion of the plate dissolved 
 
 
 
 
 f, 
 
 
 
 
 Fig. 105. — Lines in Weld Annealed. 
 
 into the welding pool reacts with the large percentage of iron oxide 
 contained therein and forms more carbon monoxide, which has no 
 opportunity to escape. Welds made on carbon-free iron do 'not 
 
 Fig. 106. — Slag Enclosed in Weld. 
 
 always appear in this particular form, as may be seen where the 
 most highly nitrogenised sections show up as dark patches not unlike 
 pearlite. These are shown under higher magnification in Figs. 99 
 and 100. 
 
THE METALLURGY OF ARC WELDING 205 
 
 Nitrogen is one of the most effective elements for making steel 
 brittle. As little as 0-06 per cent, will reduce the elongation on a 
 0-2 per cent, carbon steel from 28 to 5 per cent. Nitrogen is con- 
 tained in regular steel only in very small amounts, varying from 
 0-02 per cent, in Bessemer steel to 0-005 per cent, in open-hearth. 
 Under ordinary conditions of fusion, nitrogen has little effect on 
 iron, but under the conditions of the electric arc it becomes much 
 more active. The elimination of these nitrides and oxides must be 
 accomplished before the weld can be made ductile. Many attempts 
 have been made to do this by the alloying of various scavengers with 
 the electrode material, or by painting them on or in some way 
 attaching them to the electrode. Among other impurities that 
 may occur in a weld, the sulphur may combine with the manganese 
 present. 
 
 It is doubtful if enough sulphur will remain in the weld section 
 to do any great harm. Phosphorus forms a dangerous phosphide 
 eutectic with iron, which tends to form a brittle envelope around the 
 crystals. 
 
 Composition. — -By composition, given as the fifth item influencing 
 the quality of the weld, is meant the intentional addition of such 
 elements as nickel, tungsten, or the like. These elements in varying 
 proportions are added to steel to impart specific properties. One other 
 item which must receive consideration is the effect of overheating 
 the plate during welding. In all welds on fairly heavy sections, this 
 effect is always present, and not infrequently so weakens the metal 
 as to cause it to break just outside the weld, giving rise to mistaken 
 ideas that the weld is better than the metal welded. This overheating 
 causes a coarsening of the grain in the metal (see Fig. 103), and the 
 segregation of the pearlite into large masses enclosed in ferrite enve- 
 lopes. In general, it must be said that much depends on the opera- 
 tor, and much difficulty is experienced on this account in comparing 
 data from several sources. Great difference of opinion still exists 
 on many of these points, but the co-operation of the welding interest 
 is making rapid strides towards placing the whole art upon a more 
 scientific basis. 
 
CHAPTER XXXIII 
 BRIEF DESCRIPTION OF ELECTRIC WELDING 
 
 Electric butt welding is a method of fastening together suitable 
 pieces of metal by the creation of intense heat at a desired place 
 through the proper application of electricity, jointly with pressure 
 if butt, seam, or spot welding. The pressure must be applied — 
 if the work is to be efficient — in increasing ratio, before and after 
 the application of the electric current. The time taken to make a 
 weld varies from one-fifth of a second to four or five minutes, 
 according to the area welded. The current required is of high 
 amperage {i.e., volume), but at low voltage (i.e., pressure). 
 
 Volts x amperes = watts. 
 1,000 watts = 1 kilowatt. 
 746 watts = 1 horse-power. 
 
 The pressure is applied by hand or foot, by spring, or by hydrau- 
 lic means, according to the nature and size of the work to be welded. 
 There is no danger. The voltage or pressure used in the electrodes 
 and the exposed portion of these machines is so low — being from 
 only 2 up to, possibly, 5 volts at the most — that no one would feel 
 an embarrassing shock therefrom. It is about equivalent to the 
 voltage or pressure of an ordinary push-bell. Subsequently, heat 
 has no effect upon the weld, unless the heat is of such an intensity 
 as to remelt or burn the metal; no less degree of heat will affect it. 
 
 There are four main divisions of electric welding, namely : 
 
 Spot Welding. — This, as its name implies, is a process which con- 
 sists in welding articles together in spots instead of riveting them. 
 The spots are usually T V to J inch in diameter, seldom larger. 
 
 Seam Welding. — This is a process whereby the overlapping edges 
 of metal are joined together by fusing. 
 
 Butt Welding. --This process consists in bringing together, or 
 butting, the two ends of the metal rod, bar, etc. (not overlapping), 
 thereby causing them to fuse one into the other. 
 
 Arc Welding. — This process is used for filling in new metal along 
 the joints to be welded, the electric current melting both faces to 
 
 206 
 
BRIEF DESCRIPTION OF ELECTRIC WELDING 207 
 
 be welded at the same time as the electrodes, and filling up the joint 
 to the level of the plate. 
 
 In spot welding the electrodes on various machines may vary 
 from § to | inch or more in diameter. They should be tapered 
 similarly to a pencil, at an angle of about 45°, to a dull point. The 
 point should be slightly less than the diameter of the weld. This 
 may cause a marking or pitting of the metal where the weld takes 
 place, which can be avoided on the mild steel about 20- to 28-gauge 
 by using flat electrodes on one or both sides. On thicker materials 
 one flat electrode can be used. 
 
 Relative Cost of Riveting and Electric Spot Welding. — Riveting 
 requires, in labour, the marking off, punching, and drilling of both 
 the plates so that the holes match, and the actual operation of 
 riveting. Then there is, of course, the purchase cost of the rivets. 
 The result attained is two pieces of metal held together by a softer 
 metal. As they are held chiefly by the pinch between the two rivet 
 ends, working is likely to take place, for the rivets seldom fit the 
 hole tightly. 
 
 Spot welding saves the cost of rivets and most of the labour. 
 There is no marking off, no holes to punch, no rivets to fit in the 
 holes. The welds can be applied at a speed varying, in continuous 
 succession, from a few to over 100 per minute. In fact, on certain 
 kinds of work, 200 welds can be applied per minute. The welding 
 machines on light work are capable of making a weld every fifth 
 of a second, the speed depending upon the ability of the operator; 
 and in all repetition work the spot welding costs are 75 per cent, 
 less than riveting. 
 
 Metals Suitable few Spot Welding. 
 
 Platinum to steel or iron. 
 
 Silver to brass, steel or iron. 
 
 Aluminium. 
 
 Phosphor - bronze to self and 
 
 brass. 
 Brass to self, steel, and iron. 
 
 Copper to self, with a thin brass 
 
 insertion. 
 Steel to self, iron, and metals as 
 
 above. 
 Iron to self, steel, and metals as 
 
 above. 
 
 Cast iron is not weldable by this process. 
 
 Butt welding is principally used for repetition work, joining 
 together bar metal for crankshafts, drop forging, to stock bars, 
 tyres for cars and waggons, bands for oil drums, rings, chains, mild 
 steel to high-speed steel, etc. The scope is from wire the thickness 
 of hair to 15 inches square. 
 
 Seam welding is used for making watertight joints, where spot 
 
208 MODERN METHODS OF WELDING 
 
 welding would not suffice. Seam welding consists of welding to- 
 gether the overlapped edges of two sheets of metal by heating under 
 pressure, with copper rollers or wheels. 
 
 Usually the combined effect of the pressure and the heat causes 
 the welded seam to be nearly the same thickness as the original 
 stock. It is, however, limited in its application to thinner material, 
 and aluminium cannot be welded. 
 
 Early in 1902 there was a demonstration in Milwaukee of the use 
 of the electric arc for the cutting of steel. An enormous boiler founda- 
 tion had to be removed from the basement of a building, so heavy that 
 local mechanics despaired of being able to cut it. The electric arc 
 was requisitioned and soon cut (or, more correctly, burned) the steel 
 plate at the rate of 1 foot in five minutes, so that in a short time the 
 whole plate was divided in blocks and transported away. 
 
 Electric track welding has become an important business. 
 Rail joints are welded together just as they lie on the ties. The 
 first operation is sand-blasting to free the rail ends from dust and 
 dirt. An apparatus resembling a horseshoe is placed over the 
 rails where they join; then, strips of steel having been placed on the 
 sides of the joint, the current is turned on. The metal of the joint 
 soon rises to a welding heat. The current is next shut off and the 
 hydraulic jaws produce a great pressure which completes the weld 
 quickly. The current used is from 25,000 to 30,000 amperes at 
 7 volts. The supply at the welder is regulated at about 30 volts. 
 
 Operators often ask, What is a volt ? This is a term used to 
 represent the pressure of electrical energy. In steam we would say 
 that a boiler maintains a pressure of 100 pounds. This term relates 
 to pressure only, regardless of quantity, just as the steam pressure 
 of a boiler has nothing to do with its capacity. 
 
 An ampere is a term used to represent the current. In the case 
 of steam or water we speak of the carrying capacity of a pipe in cubic 
 feet, while in electricity the carrying capacity of wires is given in 
 amperes. 
 
 A watt is the electrical unit of power, and equals volts x amperes. 
 One watt horse-power is equivalent to lj mechanical horse-power 
 A kilowatt-hour, or k.w.h., is the electrical equivalent of mechani- 
 cal work, which would be stated in the latter in the terms of horse- 
 power. It means the consumption of 1,000 watts of electrical 
 energy steadily for one hour or any variation thereof (such as 5,000 
 watts for twelve minutes), and it is the unit employed by all power 
 companies in selling electrical power, their charges being based on a 
 certain rate per k.w.h. consumed. 
 
BRIEF DESCRIPTION OF ELECTRIC WELDING 209 
 
 K.v.a. means kilovolt amperes, or volts x amperes -h 1,000. In 
 any inductive apparatus, such as a motor or welding machine, a 
 counter current is set up within the apparatus itself. This makes 
 it necessary for the generator to produce not only amperes enough 
 to operate the motor or welding machine, but also enough in addition 
 to overcome this opposing current, although the actual mechanical 
 power required to run the generator is only that sufficient to supply 
 watts or electrical energy (volts x amperes) actually consumed in the 
 welding machine. Hence the k.w. demand of a welding machine 
 represents the actual useful power consumed for which you pay, 
 while the k.v.a. demand represents the volts x the total number 
 of amperes impressed on the welding machine -5- 1,000, to overcome 
 also the induced current set up within it. But it is the k.v.a. 
 demand that governs the size of the wire to be used in the connecting 
 up of the welding machine. K.w. divided by k.v.a. of any machine 
 is usually expressed in percentage. 
 
 According to a report submitted to the Convention of the Associa- 
 tion of Railway Electrical Engineers, and reprinted in the Railway 
 Review, December 2, 1916, ^--inch mild steel electrodes used for 
 welding 2-inch flues require a current from 60 to 90 amperes, with 
 a voltage from 14 to 16; 5-inch flues using a ^-inch mild steel 
 electrode require a current of from 110 to 140 amperes, with a 
 voltage of 16 to 20. Mild steel electrodes T % inch thick require 
 a current of from 151 to 180 amperes, with voltage from 18 to 25 
 volts. When carbon electrodes | inch thick are used for cutting, a 
 current is needed of from 250 to 370 amperes, with a voltage of 35 to 
 50 volts. In some outfits, however, carbon electrodes much smaller 
 in diameter are used, one company employing only f^-inch diameter. 
 
 When very thin sheet is welded with metallic electrodes with low 
 current values. The following data for sheet metal are based on the 
 cost of labour at Is. 6d. per hour, current at Id. per kilowatt-hour. 
 Metal No. 20 gauge, metal electrodes T V-inch diameter, current 
 10 to 25 amperes, speed 30 feet per hour, average cost f d. per foot. 
 Metal 18-gauge, metal electrodes T Vinch diameter, current 35 to 
 40 amperes, speed 28 feet per hour, average cost l|d. per foot. 
 The cost of welding ^-inch thick plates by the arc welding method is 
 about 50 per cent, of the cost of welding by oxy-acetylene on similar 
 plates. To weld plates \ inch thick by arc welding will cost 40 per 
 cent, less than oxy-acetylene. On 1-inch thick plates arc welding 
 is 15 per cent, cheaper than oxy-acetylene, but the latter is better 
 in regards to expansion owing to slow heating, which leaves small 
 granular section. 
 
 14 
 
210 MODERN METHODS OE WELDING 
 
 The electric arc claims superiority over some of the other 
 methods. First, the high temperature of the electric arc makes 
 it possible to reduce rapidly to a molten state the metal to 
 be welded. The heat being applied rapidly is not being carried 
 away from the point of the weld by the heat conductivity 
 of the metal fast enough to lower the temperature at the weld 
 appreciably. Secondly, the tools for performing the weld are com- 
 paratively easy to manipulate. The apparatus required is simpler 
 than that used for gas outfits. Thirdly, for all work, except very thin 
 materials, it is cheapest. Fourthly, the voltage of the current is so 
 low that the process is perfectly safe. If the operator is provided 
 with a proper hood or shield to protect him from the light and heat 
 of the arc, he is not exposed to any danger. The heat and light from 
 the carbon arc are much greater than that from a metallic arc. 
 
 The greatest advantage of all, probably, is that the welds can 
 be made overhead and on vertical seams by the metallic arc. The 
 arc actually carries the metal particles from the electrode into the 
 weld with considerable force, so that even with an overhead weld 
 the metal is forced clear through the space between the adjoining 
 surfaces, welding them securely. Overhead welding cannot be done 
 so easily by any other means. The welds that are most commonly 
 welded by the electric arc are mild steel, and steel castings. For mild 
 rolled steel and steel castings, electrodes or filling rods of soft iron, 
 preferably Swedish iron, are used. Tool steel may also be welded 
 with Swedish iron electrodes. Copper has been welded to steel by 
 using a copper-phosphor rod. Brass also can be welded with a 
 brass-aluminium rod, bronze with a bronze-aluminium rod. 
 
 It is not possible to weld aluminium, cast iron, or copper with 
 much success, although attempts have been made. These metals 
 are bssb lsf b to the oxy -acetylene process, with which a good weld 
 can always be made. 
 
CHAPTER XXXIV 
 
 ELECTRIC ARC WELDING 
 
 Electric arc welding is a fusion process, and as is the case in the 
 oxy-acetylene blowpipe system, the joint or weld is obtained by the 
 autogenous union of the metal. The two pieces are united by 
 filling new material between them, the electric current melting both 
 faces to be welded, while at the same time the new metal of the elec- 
 trode melts into the junction of the two. The quasi-arc process 
 is dependent upon an entirely new phenomenon brought to light 
 by investigation, and is so different in method and result from the 
 arc fusion process that it merits being put in a class of its own. 
 Not only is this new process much more rapid than any existing 
 method, but it produces a perfect joint, owing to the fact that the 
 heat introduced into the weld is automatically governed by the nature 
 of the special electrodes employed. There is no limit to the size of 
 the work which can be welded, no expensive plant or machinery is 
 required, and the current consumption is extremely light. 
 
 In this process the highly localised heating agency of the electric 
 arc is employed to bring about autogenous union. The fusing 
 starts immediately the arc is struck, but under such conditions 
 that throughout the whole operation the fused adjacent metal is 
 entirely protected from all oxidising influences. The result is 
 obtained by the use of patented electrodes in conjunction with the 
 patented method of application. The method of application is 
 rendered possible by the special character of the covering employed 
 for the electrodes, and eliminates the necessity for particular skill 
 on the part of the operator, which is an essential feature of other fusion 
 processes. Uniformly good and reliable results are obtained, and no 
 appreciable thermal disturbance is caused to the structures of the 
 surrounding metal. It is necessary, when preparing the weld, to 
 have both edges of the line of welding bevelled if it is over ^ 3 T inches 
 thick. 
 
 The importance of forming a joint which shall contain no trace 
 of oxide is so great as to deserve particular emphasis. Not only does 
 the presence of oxide greatly reduce the strength of the weld, but 
 
 211 
 
212 MODERN METHODS OF WELDING 
 
 it renders the joint peculiarly liable to corrosion. But for the 
 general welding in engineering works, shipyards, and steel foundries 
 the only requisites, beyond the electrodes, are a simple electric 
 holder, a supply of current, either direct or alternating, at a pressure 
 of about 105 volts, and a suitable resistance for regulating the current. 
 Should, however, direct current be available, this maybe used pro- 
 vided that a reactance coil is installed in each welding circuit. 
 
 The bared end of the electrode, held in a suitable holder, is 
 connected to one pole of the current supply by means of a flexible 
 cable, the return wire being connected to the work. In the case of 
 small articles the work is laid on an iron plate or bench, to which the 
 return wire is connected. Electrical contact is made by touching 
 the work with the end of the electrode held vertically, thus allow- 
 ing the current to pass and an arc to form. The electrode, still 
 kept in contact with the work, is then dropped to an angle, where 
 the arc is immediately destroyed, owing to the special covering 
 passing into the igneous state, and as a secondary conductor main- 
 taining electrical connection between the work and the metallic 
 core of the electrode. The action once started, the electrode melts 
 at a uniform rate, as long as it remains in contact, and leaves a 
 seam of metal perfectly diffused into the work. The covering 
 material of the electrode, acting as a slag, floats and spreads over 
 the surface of the weld as it is formed. The fused metal, being 
 entirely covered with the slag, is thereby completely protected from 
 all risk of oxidisation. The slag covering is readily chipped or 
 brushed off when the weld cools, leaving a bright, clean metallic 
 surface. 
 
 During the last two years much attention and investigation has 
 been carried out on the problems involved in the application of 
 the process to ship construction, with a view to the substitution in 
 a large measure of quasi-arc electric welding for riveting. The 
 investigation has been of a twofold character : firstly, by a series of 
 exhaustive tests to determine the relative strength of quasi-arc 
 welding under all conditions of stress and of various types of joints: 
 and secondly, to determine what modification of design would be 
 necessary or desirable where welding is adopted. The results of 
 these tests, set out in the following pages, establish the fact that 
 a Weld by this process is not merely as strong as, but is, in 
 fact, substantially stronger than, a riveted joint, while the various 
 modifications in design which have been evolved and patented 
 by the Quasi-Arc Company effectually overcome many difficulties 
 experienced in present practice in ship construction; and this, 
 
ELECTRIC ARC WELDING 213 
 
 coupled with a notable saving in weight of steel used. Inasmuch, 
 also, as the work of one welder is approximately equivalent to that 
 of a squad of four riveters, there should be a substantial increase 
 in the rate of joroduction, accompanied by econonw in total cost. 
 
 In order to test the strength and suitability of a welded joint, 
 it is not sufficient to be content with a mere tensile or bending test; 
 a joint which could satisfactorily pass such tests might give a very 
 poor result when an alternating or vibratory stress is applied — 
 indeed, from the point of view of ship-designers, the latter is 
 probably of greater importance than either of the former, and for 
 this reason special attention has been given to the matter. The 
 joints welded by different processes may give approximately equal 
 tensile results, but show a marked difference when subjected to 
 alternating stresses. Hence the importance of the adoption of such 
 a process of electric welding, and electrodes of such standard 
 quality, as will amount to a guarantee that a true crystal union 
 actually takes place. 
 
 The coatings may be of such a nature as to supply constituents 
 that are burnt out in the metal in welding, and so compensate for 
 their loss. In some electrodes aluminium wire is incorporated under 
 the coating. Blue asbestos yarn is specially preferred as a coating 
 for the electrode for welding mild steel and iron, as it is a reducing 
 flux and may be smeared with a composition such as sodium silicate 
 or aluminium silicate, to the very fusing temperature of the yarn. 
 Extreme care is used in preparation of the electrodes, and much 
 stress is laid on the good and regular quality of the metal of which 
 they are composed, and upon the exactness and evenness of the 
 coating. The metal electrode is positive to the work and, in fusing, 
 is deposited upon it. The coating in melting forms a vitreous slag 
 which covers the weld and flakes off more or less in cooling. The 
 slag must be carefully removed if successive layers are required. 
 The object is to protect the weld from absorbing oxygen and so 
 avoid deterioration of the quality of the metal in the weld. The 
 metallic electrode fuses into the joint prepared by bevelling the 
 edges of the pieces to be welded, so that there is not properly an arc. 
 The electrodes vary from 14 to 4 s.w.g. in diameter for ordinary 
 work up to f-inch thick. The current is direct, and the voltage 
 recommended is about 100, but much lower in amperes than 
 with Bernodos' system, varying from 20 to 75 amperes according to 
 the thickness operated upon. 
 
 Cutting is impossible with a metallic electrode. The electrode 
 melts away in the operation. If it touches the work it sticks to it. 
 
214 MODERN METHODS OF WELDING 
 
 Occasional cooling by clipping the electrode in water is necessary. 
 A carbon electrode is more easily manipulated for this purpose 
 With the metallic electrode positive to the work, welds can be made 
 upwards — that is to say, the operator can work underneath the 
 article, and weld its under surface. This requires a particularly 
 good operator. 
 
 Many tests have been made, both in the laboratory and in 
 practical service, and their utility has been fully demonstrated. 
 The work, however, must be designed to suit the process, and the 
 process must be regulated to suit the work, in order to attain 
 success. 
 
 Industry in wartime becomes founded on an entirely new 
 process: production and speed of manufacture become of first 
 importance; the cost becomes, to some extent, secondary. New 
 methods must be introduced with a rapidity unknown in peace- 
 times, and the taking of some chances becomes an absolute necessity. 
 The desirability of avoiding machine work and of reducing to a 
 minimum the labour item, the necessity of utilising unskilled labour 
 wherever possible, all become of great importance. The necessity 
 for long life is not always present; substitutes must be found for 
 many materials where a shortage exists, and margins and factors 
 of safety must be reconsidered and, wherever possible, reduced. 
 
 The steel industry with its allied and auxiliary developments 
 naturally becomes of first importance. The uses of iron and steel 
 in wartime as well as in times of peace are so manifold as to pre- 
 clude a detailed listing. In practically all the uses of steel several 
 parts must be joined together to form a whole, and in many of these 
 operations rivets have been the means of union employed. In the 
 building of ships, in the construction of all structural material, 
 wherever steel plate is used, and in places without number, the 
 rivet has been the means of union between the two separate steel 
 parts. In accordance with the law of economics, wherever a process 
 can be performed in such a way as to show an advantage in quality, 
 speed, cost, or quantity, it must supplant other methods. Electric 
 welding, in some forms, gives every promise of replacing riveting in 
 the enormous field which the latter has long held for its own. It is 
 only when a rival appears upon the field that the characteristics 
 and claims of a process are properly investigated. 
 
 Electric welding is not a new art, but in its various forms has 
 been used for many years. To those who have studied the subject, 
 the possibilities of arc, spot, butt, and other forms of welding are 
 well known. The results that can be obtained are matters of ex- 
 
ELECTRIC ARC WELDING 215 
 
 perience, and years of actual service have sufficiently demonstrated 
 the unquestionable reliability of the process. The careful and 
 elaborate scientific investigations now under way to determine the 
 characteristics and limitations of all the forms of electric welding 
 vill soon place knowledge of the art on a broad basis comparable 
 to that of our best engineering methods. The repair of the wilfully 
 damaged German ships has been one of the most spectacular 
 demonstrations of the possibility of the welding art. 
 
 In those industries in which iron and steel parts are employed 
 there is practically no limit to the opportunities for employing electric 
 welding. In salvage and repair work it is being rapidly introduced. 
 Ships, ships, and more ships was an urgent and familiar war cry, 
 yet peace does not release us from building ships. In the course 
 of a few years we may hope to see the electrically welded ship the 
 rule instead of the exception. 
 
 Arc welding is relatively slow, requiring more labour, but the 
 apparatus is lighter and more portable. Electrodes for arc welding 
 cost approximately 4d. to 6d. per pound bare, and from lOd. to 3s. 
 per pound flux-covered. There is a wide range of chemical composi- 
 tions, from almost pure iron and steel with high manganese, fairly 
 high carbon. It is certain that, where the strength and ductility 
 of a weld are important, thoroughly skilled operators are necessary, 
 and these cannot be ordinarily produced with less than six or eight 
 weeks' training. 
 
 With the introduction of electric welding in shipbuilding 
 comes the necessity of devising new methods of assembly and 
 holding the plates in position for welding. Several methods have 
 been proposed, but their success can only be demonstrated by actual 
 experience. The author suggests a powerful magnet (such as one 
 lifting iron plates in steel works) for holding the plates while they 
 are tacked. This would save bolts and the need of bolting. The 
 magnet could be moved from place to place by shutting off the 
 current. Dependable arc-welded joints can be made with an 
 average strength of over 90 per cent, of the plate strength, when 
 backed up by butt straps with an average strength of over 100 per 
 cent. Owing to the relative brittleness of arc-welded joints, much 
 fear has been expressed as to their ability to withstand long-con- 
 tinued vibration stresses and shocks. Welding affords the most 
 simple and effective means of making joints in steel plates capable 
 of holding, without leaks, the warm oil which the tanks contain 
 in service. The use of welding has lowered the cost of tank- 
 making very materially, and has reduced the amount of noise 
 
216 MODERN METHODS OF WELDING 
 
 in the tank-shops, thus making the tank-maker's job more 
 agreeable. 
 
 The apparatus required for electric welding is comparatively 
 simple and very durable. When once it is installed, the operator 
 requires only his electrodes, and a source of electrical energy. 
 A successful operator must be a man of honest temperament, con- 
 scientious, and interested to obtain the best results. He may be 
 taught in a few days to hold the arc steady ; in about three months 
 he may become an average operator. It requires some time to 
 acquire the skill necessary to produce fairly uniform results in the 
 different positions in which welding must be done. The operator 
 must acquaint himself with the flow of the metal, in order to know 
 definitely whether the current which has been selected for welding 
 is too high or too low; he must come to know if the plates being 
 welded are penetrated enough with the arc to form a good joint; he 
 must observe the movement and the condition of his work, so that 
 he can leave the least possible strain in the completed weld. These 
 and many other points are to be learned principally through 
 experience. 
 
 Preparation o£ the Work for Welding. 
 
 It is essential that the work be properly prepared before welding 
 is begun. A thorough study must be given to the job in hand before 
 anjr attempt is made to weld. This study must be first applied to 
 the effect of heat on the parts to be joined; secondly, to the accessi- 
 bility of the parts to be welded; thirdly, to the nature of the strains 
 to which the weld will be subjected; fourthly, to the cleaning and 
 assembling of the parts; fifthly, to the position in which the weld 
 can be made; and sixthly, in what condition the weld is to be left 
 when finished. 
 
 (1) The effect of heat is to produce expansions and contractions 
 which must be provided for wherever possible, otherwise severe 
 strains may be left in the plates and welds that will materially reduce 
 their effective strength or leave the work in a warped and distorted 
 condition. 
 
 (2) The parts to be welded should be made accessible, so that the 
 welding may be performed thoroughly and the work of the operator 
 simplified. 
 
 (3) A study of the strains to which the work will be subjected 
 is necessary in order to determine the kind of weld that should be 
 used. Different kinds of welds will be required according as the 
 strain is a direct tension, bending, torsion, prying, compressive, or 
 
ELECTRIC ARC WELDING 
 
 21', 
 
 a composite one ; and as the strength must be great, as in a main 
 seam, or small, as in a caulking weld. 
 
 (4) The cleaning and assembly of parts must be such as to pro- 
 vide clean, proper, and sufficient contact surface for the welded-in 
 portion, and so arranged that a good and substantial joint will 
 result. 
 
 (5) Wherever possible, the joint to be welded should be placed in 
 
 Fig. 107. — Use of Metal Electrode in Welding Steel Bands to Pressed 
 
 Corrugations. 
 
 a position which will be the least arduous for the welder. Under this 
 condition he will naturally do his best. Such a position is usually 
 in the horizontal plane. Vertical and overhead welding may be 
 done, and done well, but these positions are more difficult and tire- 
 some for the welder. 
 
 (6) Usually the " welt," or raised portion of the weld, is left on. 
 But it is sometimes necessary to remove this and have a plane sur- 
 face; for example, round the top of a tank which is to have a special 
 
218 MODERN METHODS OF WELDING 
 
 finish the entire raised portion of the weld may be removed. Under 
 these conditions a light reinforcing weld may be made on the seam 
 inside the tank to compensate for the metal that has been removed . 
 It is important, if the weld is to be made continuously in one layer, 
 to allow for a contraction of the joint as the weld progresses. Unless 
 this is done, undue warping and excessive internal strains may 
 result. The amount of this contraction varies slightly with the 
 
 Fig. 108. — Use of Carbon Electrode with Metal Filler when Welding 
 ]--Inch Thick Steel Base to the Edges of Pressed Corrugations. 
 
 speed at which the work is done, and is about 1 1 per cent, of the length 
 of the weld. Clamps are used to hold the plates the proper distance 
 apart, and these are gradually released as the weld approaches 
 them. The operator watches the opening. If it closes too quickly, 
 he hurries his welding. If it does not close quickly enough, he waits 
 for it. These precautions need not be taken in very short butt 
 welds. Such welds are too short to develop any serious strains. 
 
ELECTRIC ARC WELDING 
 
 219 
 
 Welding with Metallic Electrodes. 
 
 Many of the accompanying illustrations show samples of metallic 
 electrodes welding work in tank-manufacture. These indicate that 
 a' great variety of work can ably be done by arc welding with 
 metal electrodes, and show that such operations are thoroughly 
 practicable, and the results neat and substantial. 
 
 Fig. 109. — Seam Prepared for Hand Welding with Carbon Electrode asu 
 Operator in Position for Welding. 
 
 Welding with Carbon Electrode. 
 
 The method applied to light corrugated tanks differs from those 
 used on boiler plate tanks. The sheet steel of these corrugated 
 tanks is principally of T V and 4\ inch thick and the carbon 
 electrode is used primarily to fuse together the upturned edges 
 of the sheets. The carbon electrode is also used in conjunction 
 with a metal filler when placing the |-inch bottoms in the corru- 
 gated tanks. The metal electrode, also, is used on these tanks 
 when welding the band to the corrugations. Welding, as applied 
 to corrugated construction, is graphically portrayed in the illus- 
 trations of these tanks. 
 
220 
 
 MODERN METHODS OF WELDING 
 
 Electrodes. 
 
 The bare electrodes that have been found satisfactory for tank 
 construction are as follows : 
 
 (1) Norway or Swedish iron. 
 
 (2) Toncan iron. 
 
 (3) Armco bright hard-drawn electric welding wire. 
 (•!) Roebling bright hard-drawn electric welding wire. 
 
 These wires and the tank steel have the following analyses : 
 
 
 Steel Plate. 
 
 
 Sted W 
 
 ire. 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 
 
 Carbon per cent. 
 
 0-25 
 
 0-049 
 
 0-10 
 
 0-078 
 
 0-185 
 
 Manganese ,, 
 
 0-40 
 
 0-021 
 
 0-16 
 
 0-041 
 
 0-561 
 
 Phosphorus ,, 
 
 0-025 
 
 0-025 
 
 0-025 
 
 0-010 
 
 0-032 
 
 Silicon ,, 
 
 0-000 
 
 0-08 
 
 trace 
 
 0-000 
 
 trace 
 
 Sulphur ,, 
 
 0-028 
 
 0-08007 
 
 0-046 
 
 0-032 
 
 0-038 
 
 A satisfactory welding wire will melt and drop small particles 
 uniformly into the weld. 
 
 If there is considerable spluttering, and large globules drop from 
 the welding-rod, the weld will be very porous, and the deposited 
 metal will be poorly united to the plates being joined. 
 
 It is difficult to give universally applicable figures covering 
 amperes, speed, etc., for electric welding, owing to the effect of 
 conditions under which the work is done, the character of the work, 
 and, to a very large extent, the skill of the operator. The following 
 figures are based on favourable working conditions and a skilled 
 operator. They are approximations only, and are given merely as a 
 general guide. 
 
 Metallic Electrode Welding. 
 
 Electrode Diameter in 
 Inches. 
 
 Amperes. 
 
 i 
 
 25 to 50 
 
 a 
 
 ITT 
 
 50 „ 90 
 
 * 
 
 80 „ 150 
 
 5 
 
 125 „ 200 
 
 .1 
 TS 
 
 175 „ 225 
 
 Corresponding Plate 
 Thickness in Inches. 
 
 up to T s ¥ 
 
 itol 
 
 f and up 
 
 The same size electrode may be used with various thicknesses of 
 plate. The heavier plates will require heavier currents. Approxi- 
 
ELECTRIC ARC WELDING 
 
 221 
 
 mate speeds of welding sheet metal with metallic electrodes and 
 oxy-acetylene welding are given in the f ollowing table : 
 
 Thickness of 
 Plate. 
 
 Speed, Feet per 
 Hour. 
 
 Cost per Foot 
 
 of Electric Arc 
 
 Welding. 
 
 Comparative Cost 
 per Foot of 
 Acetylene. 
 
 i 
 
 IF 
 
 
 20 
 
 2-12 
 
 1-78 
 
 X 
 
 
 16 
 
 3-12 
 
 4-66 
 
 l 
 
 
 10 
 
 7-13 
 
 12-3 
 
 _3_ 
 
 
 6-5 
 
 12-3 
 
 36-1 
 
 1 
 
 
 4-3 
 
 19-8 
 
 much, higher 
 
 a 
 
 
 2 
 
 41-7 
 
 ;s 
 
 1 
 
 
 1-4 
 
 61-3 
 
 " 
 
 Any direct source can be used for welding, but the voltage of the 
 arc must be reduced to values of from 50 to 20. The G.E.C. have 
 
 Pig. 110. — Complete Unit eor Electric Arc Welding. 
 
 developed a special line of low-voltage generators and controls, 
 which give a very good efficiency, combined with flexibility and 
 ample protection. The generator is wound for a voltage of 60 to 
 75. In no case is it necessary to have a generator of higher voltage 
 than this. Lower voltages may occasionally be used with low circuit 
 values. The generators are usually furnished as a part of a motor- 
 generator set, although they can be supplied for a belt drive if 
 desired. The motor-generator set is the most desirable equipment 
 for several reasons : it is a self-contained unit and does not demand 
 any attention when running; the maintenance is low; the weld- 
 
222 
 
 MODERN METHODS OF WELDING 
 
 ing circuits and the shop circuits are electrically independent, so 
 that short circuits in the welding circuit will not seriously interfere 
 Avith the shop circuit; the voltage on the welding circuit can be 
 regulated, if desired, by adjustment of the generator field rheostat. 
 The control equipment consists of a main generator panel, with or 
 without a welding control circuit, with a separate auxiliary panel 
 for each operator. The equipment mounted on these panels is 
 shown herewith. In addition, there is, in series with the arc, a grid 
 rheostat for varying the current by means of 
 a dial-switch shown on the panel. 
 
 The automatic control equipment gives 
 thorough protection to the generator without 
 affecting other operators whose welding cir- 
 cuits may be connected to the same generator ; 
 this equipment consists of a protective relay 
 controlling a shunt contactor in the welding 
 circuit. The setting of the dial-switch on the 
 welding panel determines the amount of re- 
 sistance in series with the arc, and therefore 
 controls the current used. This is regulated 
 by the current required by work done. -Before 
 starting the arc the operator must set the 
 dial-switch for the amount of current required, 
 so that on starting the circuits are in noimal 
 running position. There is no necessity for 
 having any relays or switches open or closed, 
 or in any way changing or disturbing the 
 electrical circuit in order to weld. 
 
 Where welding by the carbon electrode is 
 
 to be done, thin metal can be welded using 
 
 150 to 250 amperes. Medium welding by this 
 
 process requires from 250 to 350 amperes. 
 
 Heavy welding will require 400 to 600 amperes. 
 
 Where cutting is to be done by the carbon 
 
 arc, the capacity of the set depends on the 
 
 cutting speed required. For light metal where 
 
 speed is not important, 300 amperes are sufficient, but where the 
 
 metal is 2 inches thick or more it is desirable to use heavier 
 
 currents. For this purpose up to 1,000 amperes can be used. 
 
 In addition to the equipment and accessories previously described, 
 special jobs render it desirable to have on hand other miscellaneous 
 pieces of equipment. Odd pieces of copper and carbon blocks are 
 
 Fig. 
 
 111. — Welding 
 Panel. 
 
 Setting of the dial-switch 
 determines the amount 
 of resistance in series 
 with the arc. 
 
ELECTRIC ARC WELDING 223 
 
 of much assistance as " dams" in holding the metal in place. In 
 cases where the weld must be smooth on one side, a piece of copper 
 or carbon is held against the weld, and the metal filled against it. 
 Iron and steel can be used if care is taken not to weld it. In filling 
 a hole, the bottom is often closed by holding a plate of copper or 
 carbon against it until sufficient is filled in. Care should be taken 
 to flow the molten metal against the guide-pieces, not to allow the 
 arc to play directly on them. Otherwise the weld will probably be- 
 come contaminated by this material, or else the guide-pieces may 
 be welded solid and not easily removed. A steel wire scratch-brush 
 is used to remove slight scale and rust before commencing the weld, 
 also at intervals during welding — -as when changing electrodes. 
 For small work the positive lead may be bolted to an iron plate, 
 forming the top of a work-bench. The work may be set on this 
 bench, the contact being sufficient to carry the current. In many 
 cases a vice mounted on the table will be found useful. 
 
 If the work is too large for the table it may be set beside the 
 table and a bar laid across it. This will provide sufficient current 
 carrying capacity, providing that scale and rust do not prevent 
 contact. A convenient terminal for the positive cable consists of 
 a copper hook of proper size, to which the cable is bolted. If weld- 
 ing is to be done in a room where other employees are doing different 
 work, screens should be provided around the welding operator. 
 They should be high enough to prevent the light striking a large 
 part of the ceiling, since the flicker of this light would probably 
 affect other workmen. The effect, while probably not injurious, 
 would be irritating. White walls and ceilings should be avoided 
 in a welding room. Gas-burners or annealing furnaces for pre- 
 heating fire-bricks, sand, or asbestos sheeting for covering are useful, 
 especially in cast-ironwork, which in many cases should be preheated 
 uniformly to a red heat and welded at that temperature. A recep- 
 tacle of water is desirable in which the electrode holder can be cooled 
 when it becomes too hot after continual use. 
 
 Some operators feel that gloves are necessary to protect the hands 
 from the arc. In many cases, however, the operator finds gloves 
 to be in the way, especially when working with a metallic electrode. 
 If desired, however, any leather glove will give sufficient protection 
 to the skin of the hands, which is much less sensitive than the skin on 
 the other parts of the body. The arms, face, and neck should, how- 
 ever, be covered, since exposure of these parts will probably result 
 in burns similar to sunburn, which, while not serious, are painful. 
 
 Flux. — It is the experience of a great majority of operators that 
 
224 MODERN METHODS OF WELDING 
 
 flux of any kind is unnecessary in welding. Further, that it is a 
 source of danger, in that there is liability of contaminating the 
 weld. If the work is kept clean by brushing at equal intervals, 
 and ordinary care taken in the operation of the arc, a good weld 
 can be made without flux. If these precautions are lacking, flux 
 will not make a good weld. 
 
 Preparation of Welds. — Metal that is clean is much more likely 
 to make a good strong weld. Scale, rust, grease, soot, and any foreign 
 matter will contaminate the weld. Such inclusions necessarily 
 weaken it, or else make it hard. Impurities may also make the 
 metal porous and spongy, owing to the liberation of the gases. 
 Pieces of foreign matter may prevent the molten metal from filling 
 all the parts of the weld and cause cavities. Various methods of 
 cleaning are in use: pickling for small parts, washing with petrol 
 or lye, boiling with lye and sand, sand-blasting, chiselling, scratch - 
 brushing, etc. — the method depending on the local conditions. 
 Preparatory to welding locomotive tubes to the sheets, it is sometimes 
 advantageous to send the locomotive out for a run to burn off the 
 grease, and then clean off the oxide and sootby sand-blasting. Another 
 method is to heat the boiler to normal by steam pressure, and then 
 to clean by sand-blasting or scratch-brushing. Washing with lye 
 will also remove the grease. In welding heavy sections, where it is 
 necessary to deposit several layers of metal on the surface, the 
 preceding layer should always be cleaned before starting the next. 
 
 Sections of §■ inch or less in thickness need not be bevelled, but 
 they should be separated about -J inch. Thicker sections should 
 be bevelled to give a total angle of 60° as well as separated |- inch. 
 In some special cases angles as low as 40° may be necessary, while 
 as high as 90° may be used; but an average and safe value is 
 60°. Still heavier sections may be bevelled both sides and the weld 
 made from both sides. 
 
 In the latter case a layer should be put on one side, then a layer 
 on the other, to prevent warping ; for long seams the edges should be 
 kept about |- per cent, apart, at the opposite end to which the 
 welding is started; at the end where the weld starts this is to be 
 kept open | inch. This takes some of the expansion and contrac- 
 tion of the metal in the sheet. Another method of reducing 
 contraction is to put in short sections at intervals, welding in 
 one layer at a time, starting at the centre and working altern- 
 ately to each end. Then put a layer on the open sections, and 
 continue in the same way until the weld is complete, the welded 
 section of any layer below or above the joints being broken 
 
ELECTRIC ARC WELDING 225 
 
 as in laying brickwork. Still another method: instead of be- 
 ginning the weld at the edge of the plate, start it some distance in 
 and weld towards the edge of the plate. Then a second weld is 
 started the same distance ahead of the first section. This method 
 is called back- welding. The length of each section depends on the 
 total length and may vary from 4 to 10 inches. In the welding of 
 complicated shapes, such as flywheels, some castings may require 
 preheating at certain points to produce initial expansion, which will 
 be overcome as the weld cools. In some cases the entire pieces must 
 be preheated; in others, after welding, -the whole piece must be 
 annealed. This is done by heating the pieces uniformly, then cover- 
 ing it with sand, asbestos, etc., and allowing it to cool slowly. In 
 welding cracks in plates, forgings, or castings, the crack or fracture 
 should be bevelled entirely through within -jV inch of the bottom. 
 
 In boiler work |-inch holes are sometimes drilled just beyond the 
 crack to prevent further fracture. 
 
 Welding with the Metallic Electrode. — The arc should be kept 
 short, not over |- inch in length. The current should not be greater 
 than that indicated in the table for the electrode. Excessive current 
 burnt or porous metal to be deposited. The arc should be kept 
 constant in length to ensure uniformity in the metal deposited. In 
 welding a seam the electrode should be moved with a zigzag or gyra- 
 tory motion: the motion must be an advancing one along the seam. 
 The metal will adhere only to the surface of the work actually played 
 on by the arc, so care must be taken to bring the arc in contact with 
 the whole surface to be welded. Be sure that the electrode is con- 
 nects! to the negative terminal. If the polarity is reversed the 
 arc will be more difficult to maintain, the electrode will not be as 
 good as it should be. In starting the arc, the electrode should be 
 just touched to the work, and withdrawn immediately to the required 
 distance. If the electrode is held too long in contact it will not work ; 
 in this case the relay, if adjusted properly, will operate opening the 
 circuit, after which the electrode can be knocked loose. 
 
 In welding, be sure that the arc plays over the entire surface of 
 the joint. The metal of the work is fused by direct impact of the 
 arc ; if the molten metal merely runs ahead of the arc, over the solid 
 metal of the work, it will not result in a weld. The metallic electrode 
 used is generally from 14 to 18 inches long. It may be gripped in the 
 holder, either at one end or in the middle as required by skill of the 
 operator or the nature of the work. The operation of welding over- 
 head is the same as in normal welding. The difficulty largely lies 
 in the holding of the electrode steady in the cramped position 
 
226 MODERN METHODS OF WELDING 
 
 usually required. If the arc length is kept constant the metal will 
 be successfully deposited; practice is required to accomplish this. 
 The appearance of an over weld is sometimes marred by drops of 
 metal projecting, or by uneven thickness of the deposited metal, 
 but this can be overcome by proper manipulation of the electrode. 
 A rest for the arm will sometimes assist the operator to hold the 
 electrode steady. 
 
 The Use of the Carbon Electrode. — -The holder should grip the 
 electrode from 4 to 5 inches from the end, the electrode for ordinary 
 work to be tapered to a blunt point at the working end. These 
 carbon electrodes are specially made from a superior grade of pure 
 graphite. They are stocked in three sizes — f inch, -| inch, and 
 | inch diameter — and 12 to 24 inches long. To deposit metal 
 with the carbon electrode, the arc is struck as above, but it 
 is not held long enough in one place to melt through. A pool 
 of molten metal is established, a melting-rod of metal is fed 
 into the arc melted down in the work. It should all be heated 
 thoroughly to ensure complete union before more metal is added. 
 Since a heavier current can be used with the carbon electrode 
 than with fehe metallic, faster work can be done in depositing 
 metal. The quality of the weld is not quite so good, however, as 
 when the metallic electrode is used. However, for filling holes in 
 castings,burning up worn spots, etc., the carbon weld is satisfactory 
 and should be used. Owing to the temperature and the large 
 amount of heat liberated when using the carbon electrode, the elec- 
 trode holder is liable to become very hot, and, under some conditions, 
 melt away at the end. When the holder begins to get hot it should 
 be plunged in the receptacle of water kept conveniently near the 
 operator. 
 
CHAPTER XXXV 
 SPOT WELDING 
 
 Spot electric welding is the process whereby two pieces of metal 
 are united by heating until they reach a semi-molten or plastic 
 stage, when they can easily be forced to cohere or weld by the appli- 
 cation of pressure. The complete cohesion of the heated molecules 
 makes the two pieces of metal practically solid where they are 
 forced together. It is necessary that the area to be welded should, 
 at the start, be brought into more intimate contact than the sur- 
 rounding areas, in order that the current may be properly localised, 
 and the heat generated in the region where it is needed. 
 
 Some of the advantages of spot welding are that the weld is 
 softened or changed in its texture after welding by the application 
 of considerable heat, and for this reason it can be extensively used 
 in the manufacture of stoves or other articles subject to high heat, 
 where even brazing could not stand up. There is no noise in con- 
 nection with the operation, no dirt, smoke, or wasted heat. The 
 current is only on for a brief period of the time required to heat the 
 two sheets of metal at the point of the weld, and as soon as the 
 welding is completed all expense of current ceases immediately. 
 Owing to the way the metal is forced together, no oxidation can take 
 place on the abutting surfaces; therefore, no welding compound 
 or deoxidiser of any kind is necessary. First the stock must be 
 cold rolled, hot pickled, or sand-blasted to remove all scale or dirt, 
 which acts as an insulator and cuts down the capacity of any spot- 
 welding machine. The welding machine is always ready to make 
 a joint at the will of the operator ; yet, as soon as the welding has 
 been completed, the machine is practically dead, and the current 
 expense is stopped until the next operation. The metal is in full 
 view of the operator at all times, and no smoked glasses or goggles 
 are required. 
 
 The sheets to be welded must be perfectly flat and in good con- 
 tact at the surface to be welded, so that no great mechanical pres- 
 sure is required to flatten down any bulges or dents to bring the two 
 plates of stock into good contact directly under the die-points. 
 
 227 
 
228 MODERN METHODS OF WELDING 
 
 The stock must not surround the lower horn in any way like a 
 cylinder or a rectangular box, as would be the case in welding the 
 seam-side of a can or pipe. These conditions are not to be interpreted 
 as meaning that no spot welding can be done unless they are abso- 
 lutely followed, but merely to give a basis on which the ratings are 
 calculated. If any of these conditions are violated — which is often 
 necessary, especially the last one — it will still be possible to spot 
 weld, but the capacity of the machine will be cut down. The cut- 
 down of the capacity as outlined (by the stock not surrounding the 
 lower horn) is due to the self-induction effect of the metal, which 
 tends to choke back the main current and in this way cuts down 
 the welding effect at the die-points. This is lost energy, as the 
 amount of current choked back is not used in any way. 
 
 The theory of this choking back would be too lengthy to explain 
 in full, but, to give a brief analogy, it might be compared to the back- 
 pressure effect on a petrol engine in a motor-boat, where the ex- 
 haust is located under water and the power of the engine is reduced 
 owing to the back-pressure caused by the water pushing against 
 the exhaust gases. This induction effect, so called, is only present 
 in welding iron and steel, no such effect being experienced with 
 brass. 
 
 Light gauges of sheet metal can be welded to heavy gauges or 
 solid bars of steel if the light metal is not greater than the rated 
 single sheet capacity of the machine. Soft steel and iron form the 
 best welding materials in sheet metals, although it is possible also 
 to weld sheet iron or steel to malleable iron castings of a good 
 quality. Galvanised iron can also be welded successfully, although 
 it takes a slightly longer time than clear iron and steel stock in 
 order to burn off the zinc coating before the weld can be made. 
 Contrary to common opinion, the metal at the point of the weld 
 is not made susceptible to rust by burning of this zinc, since by some 
 electro-chemical action it has been found that the spots directly 
 under each die-point and also around the point of the weld between 
 the sheets are covered with a thin volume of zinc oxide after the 
 weld has taken place, which acts also as a rust-preventative to a 
 very noticeable degree. 
 
 On spot-welded articles used in practice for some time, such as 
 galvanised road culverts, refrigerator-racks and pans, rain gutters, 
 buckets, etc., it has been found that no trace of rust has appeared on 
 the spot welds from their exposure to ordinary conditions. Extra 
 light gauges of galvanised iron below No. 28 B. and S. cannot be 
 successfully welded, owing to the fact that so little of the iron is left 
 
SPOT WELDING 229 
 
 after the zinc has been burnt off that the metal is very apt to burn 
 through and leave a hole through the sheets. Tinned sheet iron 
 or steel makes an ideal metal, giving great strength at the weld, 
 but the stock will be discoloured at this point over the area covered 
 by the die-point in operation. Sheet brass can be welded to brass 
 or steel if it contains not more than 60 per cent, copper. It is not 
 practicable to attemptto weld any bronze or alloy containing a higher 
 percentage of copper than this, as no great strength can be obtained 
 Another class of work which can be handled to good advantage on 
 a spot-welder, although it is not strictly spot welding, is the con- 
 struction of wire- work articles. 
 
 This mesh welding of two crossed wires is usually done with the 
 same two copper dies as are used for spot welding, except that the 
 dies are usually grooved in order to hold the wire in the desired 
 position to weld. The welding itself is quite as rapid as that of 
 sheet metal, but a jig to hold the wire parts together in the correct 
 position before welding the joint is usually required in order 
 to secure high production. Among common wire-work articles 
 assembled by this method of welding will be found lamp-shade 
 frames, oven racks, dish strainers, waste baskets, etc. 
 
 Spot welding requires as part of its equipment a suitable trans- 
 former. The essentials for the purpose are : 
 
 (1) Very large currents at low voltages, the currents running as 
 high as 50,000 to 75,000 amperes, the voltage 4 to 15. 
 
 (2) Different classes or thicknesses of metal having to be welded 
 by the same outfit, it is necessary to provide variation in voltage in 
 order to obtain suitable currents for the different classes of work. 
 
 (3) On account of high current it is necessary to have the trans- 
 formers as near the work as possible to avoid excessive cost of low 
 voltage bus bars. 
 
 (4) The fact that the transformer is an integral part of the weld- 
 ing machine, and as such may be subject to very rough usage in 
 factories, blacksmiths' and boiler shops, etc., or may even be ex- 
 posed to the weather, necessitates particularly rugged construction. 
 
 Spot welding is the method of joining metal sheets together at any 
 desired point by a spot, the size of a rivet, without punching holes 
 or using rivets. It is done electrically by fusing or melting the metal 
 at the point desired, at the same instant applying sufficient pressure 
 to force the particles of molten metal together. The theory is as 
 simple as its application. It is a well-known principle that a poor 
 conductor of electricity will offer so much resistance to the flow 
 of the current that it will heat, the degree of heat depending on the 
 
230 
 
 MODERN METHODS OE WELDING 
 
 amount of current and the resistance of the conductor. Copper 
 conductors carry the current with very little resistance. Place a 
 piece of iron in the circuit ; it is not so good a conductor as the copper, 
 and will heat. If the volume of the current is large, and the iron 
 conductor much smaller in diameter than copper, the iron will 
 
 Fig. 112. — Prescot Spot Welder. 
 
 quickly become hot enough to melt. An incandescent lamp offers 
 a good illustration of this principle : the copper wires leading to 
 the lamp are good conductors and remain cool; the carbon fila- 
 ment, being a poor conductor, becomes white-hot, and reaches a state 
 of incandescence. 
 
SPOT WELDING 
 
 231 
 
 Instruction for Working the Machine. 
 
 Set the regular handle to the extreme left-hand side, No. 1, 
 and the double-pole, double-throw switch to the left. Place the 
 work between the copper die-points and close the dies on the work. 
 
 Fig. 113. — This is a Specially Good Sample: First, Three Flat Bars to 
 Corrugated Sheet; Second, a Flat Plate Welded to the Flat Bars 
 and Plate; Fourth, Bent Flat Bars Welded to Corrugated Sheet, 
 Flat Bars, Plate, through All Pieces. 
 
 This will force the stock together. The current is turned on with 
 the switch. If the stock does not heat rapidly enough, turn the 
 regulator-handle to the right, or No. 2. If not enough is obtained 
 at this point, keep on until the point is reached; if enough heat is 
 
 Fig. 114. — One Piece op Angle Iron to Flat Plate. Two Spots. 
 
 not obtained, throw the double-pole switch to the right, and the 
 lever handle to No. 1. The maximum current is obtained when the 
 regulator is at the right. 
 
 There is absolutely no danger of getting a shock on the machine 
 between the upper and lower dies, as can readily be proved by plac- 
 
232 
 
 MODERN METHODS OF WELDING 
 
 ing the fingers through from the upper to the contact points and 
 turning on the currents. 
 
 The voltage is so low that it is impossible to feel anything. Do 
 not touch the wires leading from the transformer or dynamo to the 
 machine without first opening the switch on the wall. There is no 
 occasion for the operator to touch these wires in any way after the 
 machine has been connected up. 
 
 Figs. 113, 114, and 115 are samples of spot welding. 
 
 There is a limit to the thickness of sheet metal which it is 
 commercially practicable to spot weld, owing to two causes: 
 
 First, the fact that the copper rods which conduct the electric 
 current can only carry a certain quantity of current without exces- 
 sive heating. When sufficient current is carried over these copper 
 rods or die-points to bring the heavy bodies of metal up to the weld- 
 
 Fig. 115.- 
 
 -Three Pieces of Round Iron Welded to an Angle Iron. 
 Good Test. 
 
 ing temperature, the copper rods will become hot ; then they soften, 
 and the points will wear away quite rapidly. 
 
 Secondly, it being necessary to have two pieces of sheet steel 
 touching each other at the point where the weld is made, with very 
 heavy stock a slight kink or buckling of the metal will prevent the 
 flat surfaces from touching each other and making good contact. 
 Light gauges of sheet steel can be welded to heavy gauges or to solid 
 bars of steel. It is not possible to weld two pieces of cast iron, owing 
 to the crystalline structure of the metal. Sheet steel can be welded 
 to cast iron, but can easily be pulled apart. The sheet tears out 
 small particles of cast iron. Galvanised iron can be welded, although 
 it will burn off the zinc somewhat where the weld is made. The 
 author does not advise the welding of light galvanised iron above 
 24-gauge, as there is no body of metal to work on. By the time the 
 weld is done the zinc is burnt off and there is nothing left. Sheet 
 brass can be welded to sheet brass or sheet steel. There is a little 
 
SPOT WELDING 233 
 
 knack in welding work of this kind, and it may take a bit of experi- 
 menting to get the right heat and pressure. 
 
 Some grades of sheet aluminium can be spot welded, although it 
 will leave a slightly roughened surface where the die-points come 
 together. It is more difficult to weld sheet copper to sheet copper, 
 as this metal is such a good conductor of the electric current that 
 there is practically no resistance offered by the metal. Rivets of 
 any size can be heated after the rivet has been set in the rivet- 
 hole, and headed and pressed in place at one operation by use of the 
 welder. This process can be used to advantage in many cases. 
 Heat has no effect on the electric weld. For this reason the process 
 is largely used by stove manufacturers in making sheet-steel ranges, 
 and for similar work. 
 
 To facilitate the welding of awkwardly shaped articles the bottom 
 stake may be swivelled and raised or lowered. With such a com- 
 bination it is practicable so to arrange the stakes that the most 
 awkwardly shaped articles can be handled. The welder is foot- 
 operated, and no skilled labour is required. Foot pressure on the 
 pedal brings down the top electrode towards the bottom electrode, 
 pinching between them the articles to be welded, which are held 
 there by the operator. The same downward movement of the pedal 
 switches on the welding current. After welding temperature is 
 reached — judged by the colour — further pressure on the pedal trips 
 the switch and applies pressure to force the heated metal into 
 welding. The pedal is made reversible, so that the welder may be 
 operated by either right or left foot. The tips are made of copper 
 and are water-cooled, with a constant flow of water. The system 
 of c'ooling is so efficient that the tips may be touched by hand 
 immediately after welding. Usually the tips last for a few weeks' 
 constant use. They are easily and cheaply replaced. 
 
 Operating Instructions. — Select the pieces to be united, range 
 them together in the required position, just as they will be when 
 welded, and clamp them thus between the electrode tips. This is 
 done by holding the pieces on the bottom electrode until the top elec- 
 trode has been brought down by depressing the foot-lever. Directly 
 the pieces are clamped between the electrodes, they become white- 
 hot at the points where the electrode tips make contact; and the 
 pressure between the surfaces, maintained by the foot-pedal, forces 
 the metal to unite and forms a weld. The metal should be as free 
 as passible from rust, scale, dirt, or other foreign matter likely to 
 hinder the passage of the welding current. No preparation is neces^ 
 sary unless the metal is very dirty, in which case it will be economical 
 
234 MODERN METHODS OF WELDING 
 
 to clean it, since less current will be required. The time taken to 
 weld varies from a fraction of a second to perhaps one second. No 
 definite rule can be given that will decide all cases. It depends on 
 the thickness of the metal, its freedom from dirt, the position of the 
 plug in the plug-box, and the diameter of the spot weld made. 
 A few hours' experimenting is generally sufficient to teach the average 
 operator. The plug should be in No. 1 hole when welding the 
 thickest material, in No. 4 for the thinnest. The shape of the elec 
 trode tip decides the diameter of the spot. If a small spot be 
 required, the tips must be reduced at the tip ; if a larger spot, the 
 tips must be flattened. 
 
 Adjustments, Tips. — The electrode tips will require to be filed 
 from time to time, to remove any metal which may adhere to them. 
 Each tip is quite easily removable after loosening with a spanner. 
 Care should be taken not to turn the tip completely round, as one 
 would a nut. Just move it from left to right a few times, when it will 
 be loosened and may easily be removed. 
 
 Three types of electrode tips are made : 
 
 (1) Concentric — -that is, with welding point exactly in the middle 
 of the electrode. 
 
 (2) Eccentric — that is, the welding point a little out of centre. 
 
 (3) Flat — that is, no welding point at all. 
 
 The concentric type is most often used, but it demands a maxi- 
 mum amount of clearance around the weld. For welding in corners 
 and places where there is little room, the eccentric is used. The 
 flat type may be used with either a concentric or eccentric tip, but 
 it is only used when it is desired to avoid, on one side, the little 
 indentation made by the pointed electrodes. All these tips are 
 interchangeable, suitable for either top or bottom stakes. 
 
 Pressure while Welding. — One of the most important points to 
 remember is the necessity of having a fair pressure between the sur- 
 faces to be welded before the current is switched on by closing the 
 switch at the back of the welder. This can easily be arranged for 
 by moving the top or bridge of the switch up and down the rod on 
 which it is mounted, adjusting so that the electrode tips are forced 
 well together before the switch closes. Care must be taken to 
 see that the electrode tips are directly in line when touching. If 
 one is a little to the side of the other, the weld is likely to be burnt. 
 Directly the article to be welded is clamped between the electrodes, 
 the secondary circuit is closed, but no current flows until sufficient 
 pressure has been set up by the foot-pedal to close the primary switch 
 at the back of the welder. Closing the secondary circuit before the 
 
SPOT WELDING 235 
 
 primary ensures a steady flow of current through the point of the 
 weld, free from sparking. 
 
 Switch. — This is fitted at the back of the welder and should be 
 inspected from time to time to ensure the points of contact being 
 kept clean. 
 
 Hinge. — To permit the top arm to move down easily when the 
 foot-pedal is depressed, while yet maintaining good electrical con- 
 tact, it is carried in a special hinge. The arm rocks up and down 
 on ball-bearings, and electrical contact is ensured by two discs, 
 which are a sliding fit in the secondary casting, and are pressed up 
 against the top arm by the springs under the nut-washers on each 
 side. These nuts may require tightening from time to time, so as 
 to keep the contact discs in close contact with the top arm. Great 
 care must be taken not to make them too tight, or excessive friction 
 will be set up and the faces will be scored. No lubricant must be 
 used. Should the faces at any time become scored, take the hinge 
 apart, smooth the faces, and polish with metal polish. The ball- 
 bearings require no attention. 
 
CHAPTER XXXVI 
 ELECTRIC BUTT WELDING 
 
 Electric butt welding is a process wherein two pieces of metal are 
 united by the cohesion of their molecules, induced by the application 
 of pressure when they are in a plastic or molten stage through being 
 heated by some process. In the process of electric welding the heat 
 is induced by passing a large volume of electric current at a low 
 pressure through the two surfaces of the pieces to be welded, the 
 heating effect in any electrical circuit being evoked by the resistance 
 of the metal to the flow of the current. When the point between 
 the abutting ends, which has the highest resistance in the circuit 
 and therefore the highest heat, has reached the proper welding 
 temperature, the current is turned off and the pressure applied 
 mechanically to force the molten ends together, thereby producing 
 a weld. 
 
 Butt-welding machines are designed for the manufacturer who 
 has a large quantity of this kind of work to do, where there are many 
 pieces of one kind to weld. They are not intended to replace the 
 blacksmith for general repair work where there are a few pieces 
 of various sizes to be welded. It is purely a production proposition 
 for a volume of work with a minimum of cost. 
 
 In butt welding the two pieces of metal are placed in the clamp- 
 ing jaws of the machine with a proportion of the ends extending 
 beyond the jaws. The electric current is turned on by means of a 
 switch, and the abutting ends of the metal instantly begin to heat. 
 The operator quickly learns to judge by observation when the weld- 
 ing temperature is reached. When he sees that the metal is hot 
 enough, he applies the pressure and forces the two metal ends of the 
 pieces into each other, at the same time turning off the current, and 
 the weld is made. The metal is in full view of the operator all the 
 time, instead of being hidden by the coal and flame in a forge fire. 
 No smoked glasses or goggles are needed any more than they are by 
 the blacksmith. There is no scarfing to be done, and owing to the 
 way the metal is forced together, there is no oxidation such as there 
 would be in the open fire. Consequently, no welding compound is 
 necessary. In a forge fire a thin film of oxide forms on the metal, 
 
 236 
 
ELECTRIC BUTT WELDING 237 
 
 which must be removed by a welding compound from the two sur- 
 faces to be joined before a good weld can be made. 
 
 With electric welds the heat is first develoj)ed in the centre of 
 the metal. In consequence, it is welded there as perfectly as the 
 surface. When welding electrically, little energy or heat is wasted 
 in heating more of the material on either side of the weld. The 
 operator has complete control of the electric current by means of his 
 current regulator and switch. He can quickly obtain any heat 
 desired, from a dull red to the melting-point of the metal. The 
 instant the weld is made the expense for current stops. Owing to 
 the low voltage employed across the work itself there is not the 
 slightest danger of injury to the operator. He cannot even feel the 
 current if he should come in contact with it across the dies. The 
 parts to be welded can be kept in fairly close alignment by the 
 clamping of the jaws, which can be given almost any shape desired 
 to hold the work. The current has no effect on the welded metal, its 
 action being to heat the metal. The copper clamping dies are good 
 conductors, and a bar of iron, being comparatively a poor conductor, 
 when placed between the clamping dies of the welder becomes 
 heated in attempting to carry the large volume of current. The 
 degree of heat depending upon the amount of current and resistance 
 of the conductor when the ends of the two pieces of bar are brought 
 together, this is the point of the greatest resistance in the electric 
 circuit, and the abutting ends heat most rapidly. 
 
 Production will depend largely on the operator, the size and 
 shape of the piece to be welded, and the kind of machine used. 
 There is a wide range in the time between the heavy pieces and the 
 light pieces which can be handled rapidly and easily. Some of the 
 smaller machines deal with several thousands of pieces a day, and 
 to get the maximum output the best machines should be adopted. 
 The welding machine can be used on one phase of a three-phase 
 system, but cannot be connected to more than one phase of a three- 
 phase circuit. Direct current cannot be used, because there is no 
 way of reducing the voltage without interposing resistance, which 
 uses up the power. The voltage used at the weld is from 1 to 15, 
 depending on the size of the welding machine. To obtain this 
 low voltage or pressure, a special transformer inside the machine 
 reduces the power line voltage down to the range. It is like 
 reducing steam pressure from 100 to 10 pounds. 
 
 The welding transformer which produces the heavy current across 
 the work is supported with the frame. Single-phase alternating 
 current is taken from a generator, or power circuit, and is stepped 
 
238 MODERN METHODS OE WELDING 
 
 down by the transformer to a low potential of from 1 to 15 volts. The 
 secondary winding of the transformer is connected to the platens, 
 and the current travels through the platens, clamps, and metals 
 to be welded, thereby completing an electric circuit. Since the 
 current value rises as the potential falls in the secondary circuit, 
 and since also the heating effect across the work is directly propor- 
 tioned to the current value, it is easily seen why a transformer is 
 necessary to produce a heavy current by lowering the time potential. 
 Owing to the intermittent character of the load, there is no standard 
 rating for welder transformers. Different makers will give entirely 
 different ratings for their machines, and not infrequently make 
 misleading statements regarding the current used. Regardless of 
 the rating in k.w. capacity, there can be very little difference in the 
 actual amount of current consumed unless an exceptionally bad 
 transformer design is used. To heat a given size stock to the 
 welding temperature in a given time requires approximately an 
 invariable amount of current. 
 
 The following illustration shows a machine for welding tool steel. 
 In any tool welding there are different kinds of welds to be made, 
 which require different classes of dies and two different types of 
 machines. We will first deal with the class of work which comes 
 under reamers, milling and key-way cutters with taper shanks, 
 and other similar make-up. The high speed and the carbon steel 
 pieces should be prepared to secure the best results on a production 
 basis. When making this style of tool the dies should be specially 
 prepared to do the work. Since the high-speed steel has a higher 
 resistance than the carbon steel it has a great tendency to reach the 
 plastic stage of heat sooner than the latter. Eor this reason it should 
 have a shorter projection beyond the dies to secure still greater 
 cooling effect and to retard its heating as much as possible. 
 
 Thorough tests have been made on the strength of electrically 
 welded bars which prove that they are almost as strong at the welded 
 joint as at any other cross-section of the metal. When welding in a 
 forge, the outer surface is heated first, and the inner part does not 
 very often reach welding heat, the result being an imperfect weld. 
 The only preparation of the stock necessary for this process is that, 
 when it is rusty or covered with blue scale, the rust and scale 
 should be removed sufficiently to give good contact of clean metal 
 at the gripping dies, as both scale and rust are poor conductors. 
 
 The butt-welding process is applicable to the welding of pieces 
 having practically the same cross-section at the joint. A very few 
 seconds after the current is turned on in the welder, the metal reaches 
 
ELECTRIC BUTT WELDING 
 
 239 
 
 a white heat, and is in a partially molten state. By means of heavy 
 pressure the ends of the metal are forced into each other in this semi- 
 
 Fig. 116. — Butt-Welding Tool Steel to Mild Steel Bars. 
 
 fluid condition, extruding all burnt metal, thus making a homogene- 
 ous mass and a perfect weld. A projection or fin will be raised where 
 the ends come together, by the squeezing out of the burnt metal, 
 
240 MODERN METHODS OF WELDING 
 
 which may be very slight on an "upset," or quite a fin may be 
 raised in a flash weld. Both are shown below. 
 
 Butt Electric Welding Process. — The parts to be welded are 
 brought into contact under pressure and then a current of high 
 amperage under low voltage is passed from one part through the 
 joint to the other part. Because of the high resistance at the con- 
 tact areas, the metal at the joint is quickly brought to a welding 
 heat, when the plasticity of the metal allows the pressure to cause 
 movements of the two parts towards each other. Under the combined 
 temperature and pressure the parts are welded, much as welding 
 is done under the blacksmith's hammer — perhaps, above all, in 
 that the heating and welding operations are performed practically 
 at the same time and almost instantaneously. As soon as the weld- 
 ing heat is reached the welding is immediately effected. In the 
 lighter kind of work the rapidity with which this heat is attained is 
 
 Fig. 117.— Butt Weld. Right, Flash Weld; Left, Upset Weld. 
 
 quite remarkable. Although the process is not applicable to every 
 form of welding, yet the sphere of its utility is very wide, and 
 the quality of the work effected by it is unquestionably good. 
 It is necessary for its effective operation to have at command heavy 
 flows of current, but, on the other hand, the voltage is very low. 
 In some cases as low an electromotive force as half a volt is all that 
 is required. In actual practice from 4 to 6 volts is about the highest 
 pressure worked with. The temperature of the metal to be welded 
 is raised simply by a very heavy current flowing through a restricted 
 area. The British Insulated and Helsby Cables, Ltd., make a large 
 variety of these machines for electric welding on the resistance 
 system. The electric arrangements involved in all machines are 
 fundamentally the same as those employed in all systems of resist- 
 ance welding — that is to say, each machine embodies a transformer, 
 wound so as to perform the particular work desired under the con- 
 ditions of voltage and periodicity of the supply current available. 
 The secondary coil of the transformer consists of a single convo- 
 lution, having a large cross-section of copper, which terminates ex- 
 
ELECTRIC BUTT WELDING 
 
 241 
 
 ternally in the two electrodes, which are of various forms. The pieces 
 to be welded are brought between these two electrodes, thus com- 
 pleting the electrical circuit of the secondary coil, so tr at when the 
 primary circuit is closed a heavy current flows in the former, as the 
 resistance to the flow of that current is practically all centred in the 
 surfaces in contact. Since the ohmic resistance of the secondary 
 
 Fig. 118. — Butt Welder for Tool Steel and Other Work. 
 
 winding is comparatively negligible, great heat is developed between 
 the electrodes, and the material between them is cpiickly brought up 
 to a welding temperature. 
 
 Attention may be drawn to the various types of butt-welding 
 machines to suit different purposes: (1) Welders for wire with auto- 
 matic upsetting gear for uniform section iron, steel, or non-ferrous 
 
 16 
 
242 
 
 MODERN METHODS OF WELDING 
 
 metals; (2) welders for manufacturing purposes with hand or auto- 
 matic upsetting gear for regular sections; (3) chain welders. 
 
 Wire welders is the term which applies to the machine in No. I, 
 some capable of welding 0-024 diameter. Larger machines of this 
 type are able to weld material up to 1 inch square. The machines 
 in the second category are used for such manufacturing purposes, 
 amongst a host of others, as welding pipe refrigerator-coils, milk-can 
 rings, perambulator rings, printers' chases, fittings to casement 
 
 t^2^gSjK& r "-if X '"- ' '"^^SSHlMfi 
 
 Fig. 119. — Butt Welder making Chains Automatically. 
 
 frames, carriage and coach work parts, trellis work, coupling links, 
 brake rigging, travelling-bag frames, low-grade shanks on high-speed 
 tools, drills, taps, etc. 
 
 The chain welders form a class by themselves, though the general 
 principles involved are very much the same as those of other 
 machines. The manufacture is carried out from coils of wire by two 
 machines, the first of which bends the links and threads them into a 
 chain, whilst the second forms the welds. Although the first machine 
 is not shown in this book, it is necessary to describe it, since the 
 
ELECTRIC BUTT WELDING 243 
 
 complete process cannot be properly understood without it. Three 
 machines are made which deal with wires from /^ to ■£$ inch. The 
 smallest machine turns out 50 to 60 links per minute and takes 
 1 to 2| horse-power to drive it. The middle size turns out 40 to 50 
 per minute, and requires 2 to 4 horse-power to drive it. The large 
 one makes from 20 to 30 links per minute, and the horse-power 
 needed is 3 to 7. The minimum proportions of the links made on 
 these machines are — -length 5 diameters and width 3 diameters 
 of wire. 
 
 General Information. — The material to be welded should be ground 
 or filed flat and square at the abutting ends, otherwise accurate 
 results cannot be obtained. The wires to be joined are each gripped 
 in a vice and the two ends projecting equally; one of the vices is 
 movable and the other fixed. While the machine is being set up a 
 spring pressure is taken up by a pawl engaging with a rack. While 
 welding, the pawl is disengaged, and this pressure is transmitted to 
 the joint. As long as the wires are cold the side remains stationary, 
 but as soon as the current is sent through they soften and give way : 
 the weld is jumped as soon as the required temperature is reached, 
 and simultaneously the current is cut off, nothing further taking 
 place. 
 
 The illustration on the opposite page is of a chain welder, which 
 has been described previously. 
 
CHAPTER XXXVII 
 
 ELECTRIC SEAM WELDING 
 
 Seam welding is a process of joining two overlapping edges of sheet 
 metal for their entire length by perfect cohesion of the molecules 
 of the material itself, without the application of any solder or spelter 
 between the edges of the joint. In the process of seam welding the 
 heat is produced by passing a large volume of electric current across 
 the joint of the edges to be welded by the employment of a copper 
 roller on one side of the joint, a copper track or horn underneath. 
 In any electrical path, wherever high resistance is interposed, heat- 
 ing will result. The higher the resistance to the current the greater 
 will be the heating effect. In seam-welding machines, since the 
 copper rollers and horn are good conductors, the joint between the 
 edges of the metal to be welded is the point of highest resistance, 
 and it is evident that the greatest heating effect will be at this point. 
 As the roller passes over the joint, heating the stock to a plastic 
 state beneath it, pressure is simultaneously applied by the springs on 
 the roller to force the edges together as fast as they are heated. 
 
 Since 20-gauge metal and lighter heats very readily, the pressure 
 and the heating can be effected at the same instant of contact by the 
 roller. It is possible to weld as fast as 6 inches a second. The 
 only preparation necessary for seam welding is that the stock must 
 be absolutely clean — -that is, free from any traces of rust, scale, 
 grease, or dirt — if a tight, neat joint is desired. If it is not necessary 
 for the joint to be tight, the stock need not be so clean, although 
 heavy rust and scale will prevent the carriage of the full current, 
 the heating will be affected, and the weld Mali not be so good. 
 
 In welding sheet brass from 22- to 30-gauge, to secure a perfect 
 joint the metal should be carefully pickled and washed to remove all 
 traces of grease and tarnish, which tend to prevent the passage of 
 the current across the joint of the edges. The metal should be 
 welded soon after pickling, as, no matter how carefully it may have 
 been washed, oxidation is always sure to start very shortly after the 
 brass has been removed from the pickling acid. 
 
 Steel to be successfully seam welded should not have a carbon 
 
 244 
 
ELECTRIC SEAM WELDING 
 
 245 
 
 content of over 0-15 per cent. A higher carbon steel than this has a 
 tendency to crystallise at the point of the weld, owing to the rapid 
 cooling of the welded portion from the surrounding cold metal. After 
 welding, the joint will be found to be about one-third thicker than the 
 thickness of the metal. It is possible by applying more pressure to 
 reduce this finished thickness, but it wears more on the copper roller 
 to do so. In seam welding brass, a soft, annealed metal should be 
 used, for, although hard rolled brass can be welded, it forces the two 
 edges together very much, and the finished joint under these con- 
 ditions is almost twice the original thickness. With a soft, annealed 
 
 Fig. 120. — Electric Seam-Welding Machine. 
 
 brass the finished joint will not be over a third greater than the single 
 metal thickness, and by applying sufficient pressure it can be reduced 
 to not over 10 per cent, thicker. 
 
 The principal advantage of the process of seam welding in brass 
 and other non-ferrous metals is that no spelter or flux is required, 
 nor is there any volatilisation of the zinc, the metal itself furnishing 
 its own cohesive properties. This allows of great speed in production. 
 The great ability of a seam welder to secure the highest production 
 lies, not only in its welding qualities, but in the adaptation to the 
 welding machine of a suitable jig. The jig holds the work properly, 
 
246 MODERN METHODS OF WELDING 
 
 and also enables the operator to place the piece in it and remove 
 the same in the shortest possible time, since the welding itself is 
 very fast compared with any other method of making a continuous 
 joint. 
 
 Fig. 120 is a photograph of a seam-welding machine. The 
 operation is very simple, once the machine is set up, for any given 
 piece of work for which a special jig has been built. After placing 
 the piece in the jig and locking it there securely, the operator de- 
 presses the foot-pedal, which throws in a clutch and starts the copper 
 roller across the work. By the proper setting of adjustable control- 
 stops on the control rod on the top of the machine, the current is 
 automatically turned on as the rollers enter on to the overlapping 
 edges of the piece to be welded, and is automatically turned off when 
 the roller reaches the end of the stroke. Another stop reverses the 
 travel of the roller, bringing it back to its starting position. The con- 
 trol stop may be adjusted to turn the current on and off at any point 
 along the roller, for doing a seam shorter than the maximum of the 
 welding machine. The roller stroke may also be shortened so that 
 a complete cycle of operations will be accomplished in the shortest 
 space of time on seams shorter than the maximum seam capacity of 
 the machine. In order to keep the copper roller from overheating 
 in action, water is introduced through its bronze bearings on- each 
 side. This same water circulation passes also through the under 
 copper horn or madrel, then through the cast copper secondary of the 
 transformer, so that the machine can be operated continually — 
 twenty-four hours a day if desired — without overheating. 
 
 Most seam-welding machines are equipped with variable- speed 
 motors in order to give a range of variation in roller travel speed, 
 which is necessary for different lengths and thicknesses of stock. 
 They are also equipped with a current regulator to give fifty different 
 voltages at the copper roller. 
 
 To effect good sliding contact with copper track, several springs 
 are employed on each side of this slide, which carries the copper 
 rollers. The lower horn is bolted directly to the lower terminal of 
 the transformer secondary. .The particular design in each case will 
 depend upon the size and the nature of the work. 
 
CHAPTER XXXVIII 
 EYE-PROTECTION IN IRON WELDING OPERATIONS 
 
 In welding operations three kinds of radiations must be guarded 
 against, one or all of which may be present to an injurious degree. 
 The problem is to j)rovide a perfectly safe filter that will permit of 
 the greatest degree of visibility, and at the same time will exclude 
 the infra-red, or heat, rays and the ultra-violet rays. Ordinary glass 
 lenses of special colours or combinations of colours are required. 
 I show the spectra of a number of commercially available glasses and 
 combinations of these glasses, and a glance at these charts will 
 show what arrangement of filter will provide the best protection 
 against the radiations of the welding arc. 
 
 Radiation from an intensely heated solid or vapour may be 
 divided under the headings: (1) Invisible infra-red rays; (2) visible 
 light rays; (3) invisible ultra-violet rays. 
 
 There is no clear line of demarcation between these divisions, 
 as they melt gradually one into the other like the colours of the 
 visible spectrum. When the heated matter is solid, such as the 
 filament of an incandescent lamp, the visible spectrum is usually 
 continuous — -that is, without lines or bands, but when it is in the form 
 of gas or vapour, as in the iron arc used for welding operations, the 
 spectrum is divided up into bands, or is crossed by lines which are 
 characteristic of the element heated. 
 
 In Fig. 121 A shows the continuous spectrum made by the light of 
 a Mazda lamp operated at normal voltage, and is the line of spectrum 
 made by a disruptive arc between iron terminals. 
 
 If A and B (Fig. 121) were coloured they would show all the hues 
 of the prismatic spectrum from red at the left to violet at the right, 
 as roughly indicated by the vertical dividing lines. The iron spec- 
 trum B falls a little short of the continuous spectrum A in the red, 
 but it is more intense than A in the visible blue and violet, and it 
 also extends farther into the ultra-violet. The spectrum B contains 
 many lines besides those pertaining to iron, principally those of 
 carbon, nitrogen, and oxygen, these elements being unavoidable 
 components of the electric spark discharge. Inspection of A and B, 
 however, will serve to indicate the extent and general characteristics 
 
 247 
 
248 
 
 MODERN METHODS OF WELDING 
 
 of the visible light that is emitted by highly treated iron vapour 
 in the process of arc welding. 
 
 The radiations under the foregoing three headings, although of 
 
 
 vioiet 
 
 Ultra 
 Violet 
 
 Fig. 121. — Spectrum of a Mazda Lamp; Spectrum op Iron Arc. 
 
 common origin, produce very diverse effects upon our senses. Thus, 
 the infra-red rays produce the sensation of heat when they fall on our 
 unprotected skin, and, therefore, special glasses are required to pro- 
 tect the operator from their harmful effects. 
 
 Fig. 122. — Pfund Gold Glass Goggles. 
 
 For welding with acetylene and for light electric welding it 
 may be necessary only to protect the eyes with goggles fitted with 
 suitable coloured glasses. Fig. 122 shows a good form of goggles 
 
EYE-PROTECTION IN IRON WELDING OPERATIONS 249 
 
 which are fitted with lenses of pfund gold glass to which reference 
 will be made later. 
 
 Fig. 124 illustrates the front and back views of a hand shield, 
 which is made of light wood and has a safety coloured glass window 
 in the centre. This device is used for medium weight electric welding 
 work which can be done with one hand, and it serves the double 
 purpose of protecting the eyes of the operator and shielding his face 
 from the heat rays and the ultra-violet radiation which would other- 
 wise cause a severe sunburn effect. 
 
 For heavy electric welding which requires the use of both hands 
 it is common practice for the operator to protect his eyes and neck 
 
 1 
 
 ^POTHpPV 
 
 
 
 >-• .. • ") 
 
 ■ 
 
 1 
 
 
 "^B 
 
 Fig. 123. — A Popular Form of Helmet with Circular Window. 
 
 with a helmet fitted with a round or triangular window of safety 
 glass. These helmets are usually made of some strong, light material 
 such as vulcanised fibre and are designed so that they can be slipped 
 on and off easily, the weight resting upon the shoulders of the opera- 
 tor. A useful form of helmet with a circular window is shown in 
 Fig. 123. Front and back views of another form of helmet are seen 
 in Fig. 125. 
 
 It therefore naturally follows that a much clearer definition of 
 an object is obtained by combination of yellow-green light than by 
 red alone, or especially by blue or violet light alone. The eye is 
 also more sensitive to the yellow and green rays than to the red and 
 
250 
 
 MODERN METHODS OF WELDING 
 
 blue rays, or, in other words, yellow-green light has the highest 
 luminous efficiency. This may easily be verified by looking at a 
 sunlit landscape or fleecy clouds in a blue sky through plates of 
 different coloured glass. A glass of a light amber colour, slightly 
 tinted with green, will clearly bring out details that are hardly 
 observable without the glass, and which can be obscured entirely 
 by a blue or violet glass. It is therefore obvious that, in order to 
 
 obtain the clearest definition or 
 visibility with the least amount of 
 glare, the selection of the colour tint 
 in safety glasses should properly be 
 decided by an expert, but the depth 
 of tint, or, in other words, the 
 amount of obscuration, may be best 
 determined by the operator himself 
 owing to the individual difference 
 in visual acuity which will permit 
 one man to see clearly through a 
 glass that would be too dark for 
 another man. 
 
 A proper selection of colour 
 tints can be assisted by spectro- 
 scopic examination, and the various 
 spectra shown in the accompanying 
 photographs are presented with this 
 purpose in view. 
 
 Fig. 126 shows different spectra 
 made by transmitting the light of a 
 Mazda lamp operated at normal 
 voltage : C, through clear colourless 
 glass; D, through ruby glass. 
 
 The screen of clear colourless 
 glass in C naturally transmits all 
 the colours of the visible spectrum, 
 extending from the extreme red 
 to the extreme violet and penetrating slightly into the ultra-violet, 
 because the latter rays, although they are not visible to the eye, 
 are highly actinic, and therefore affect the photographic plate. In 
 this case, however, we only see just the beginning of the ultra-violet 
 spectrum, as the glass plate and the glass prism of the spectroscope 
 absorb and cut off all but a few of the least refrangible ultra-violet 
 rays. 
 
 Fig. 124. — Welder's Hand Shield. 
 
EYE-PROTECTION IN IRON WELDING OPERATIONS 251 
 
 The ruby glass, used as a screen in D, transmits all the red and 
 orange rays with a trace of the yellow, but it absorbs and cuts out 
 all other colours. 
 
 The glass used in spectrum E is made by the Pittsburg Glass 
 Company (Pa.), and is termed " Belgian pot-yellow " glass. It cuts 
 off a little of the red, transmits all the orange and yellow rays and a 
 portion of the green, but cuts out all the blue and violet. 
 
 The emerald-green glass marked F is seen to transmit all the 
 yellow and green, with a considerable portion of the red and orange 
 and also of the blue. 
 
 The spectrum made through the cobalt-blue glass marked G, 
 shows the transmission of a band of red and a band of yellow- 
 
 Fig. 125. — Front and Back Views of Thin Sheet Aluminium Helmet Sup- 
 ported by a Head Band and Fitted with Rectangular Opening. 
 
 green, but it is chiefly marked by its strong transmission of the blue 
 and violet, and especially in its being a little more transparent to 
 the ultra-violet than the colourless glass A. 
 
 These five glasses are samples taken from actual service, but on 
 account of the fact that all coloured glasses are subject to con- 
 siderable variation in tint and depth of colour, caused by differences 
 in chemical composition, heat treatment, etc., the spectra shown in 
 Fig. 126 can be considered as only generally representative; samples 
 of blue glass, for example, have been tested and found to absorb 
 very much more of the red, yellow, and green than the sample 
 shown in G. 
 
 In H is seen a representative spectrum taken through a noviweld 
 
252 
 
 MODERN METHODS OF WELDING 
 
 glass (No. 6 grade), which presents an excellent colour combination 
 to secure clear definition with the least amount of glare. 
 
 It is possible to produce satisfactory colour tints for welders' 
 glasses by combining plates of different coloured glass. The results 
 of some of these combinations are shown in the spectra of Fig. 127, 
 which were made with the same source of light as those of Fig. 126. 
 
 In Fig. 127 J shows the full spectrum through clear colourless 
 glass for comparison, the same as C in Fig. 126. 
 
 In K we see the effect of combining yellow and blue glass (E and 
 G of Fig. 126), which combination makes a fair resemblance to novi- 
 weld, and is giving satisfactory service in certain work where the 
 cost of noviweld prohibits its use. The tint of this combination 
 
 Infra fed and Ve How 
 
 f?ed lor&noe Liiv*6r*en| 
 
 Slue and 
 Violet 
 
 Ultra 
 Violet 
 
 Fig. 126. — Sundry Spectra 6. 
 
 E, Through "Belgian pot-yellow glass"; F, through emerald-green glass; 
 
 G, through cobalt-blue glass; H, through No. 6 " noviweld " glass. 
 
 is inclined rather too much to the red, and is somewhat weak in the 
 yellow-green, but these defects could be largely overcome by a care- 
 ful selection of the plates. 
 
 The spectrum L in Fig. 127 results from a combination of ruby 
 and emerald-green glass (D and F in Fig. 126), which has been found 
 satisfactory for certain work and is now used extensively. 
 
 The result of combining ruby and blue (D and G in Fig. 126) is 
 shown by M in Fig. 127. It was formerly used to some extent, but 
 is now almost universally superseded by L. 
 
 The spectrum N was taken through a single plate of noviweld 
 (No. 5 grade), which presents the elements of an ideal colour com- 
 
EYE-PROTECTION IN IRON WELDING OPERATIONS 253 
 
 bination, being weak in the red while transmitting all the orange, 
 yellow, and green, but totally excluding the blue and violet. 
 
 The spectrum P was taken through a piece of amber mica having 
 a little darker tint than No. 5 noviweld. Its close resemblance to 
 the noviweld spectrum is remarkable, and if it were possible to 
 procure a clear dark amber mica in pieces large enough to be service- 
 able, this material, when protected from mechanical injury between 
 plates of plain clear glass, would closely rival the noviweld. Clear 
 dark amber mica of uniform tint and even cleavage is, however, 
 very difficult to procure, for which reason there is no probability 
 that it will ever supersede glass for protective purposes. 
 
 In selecting coloured glasses, great care should be taken to dis- 
 
 Fig. 127- — Sundry Spectra 7. 
 
 card all samples that show streaks or spots, as these defects are liable 
 to produce eye-strain. The glass should be uniform in colour and 
 thickness throughout, and the coloured plates should be protected 
 from outside injury by a thin piece of clear glass that can easily be 
 renewed. 
 
 Having considered briefly the best means for toning down the 
 glaring and flickering visible light produced in the welding process, 
 we may now direct some attention to the infra-red and the ultra- 
 violet rays, which always accompany the visible glare. 
 
 When the invisible infra-red rays encounter any material which 
 they cannot penetrate or which is opaque to them, they are absorbed 
 and changed into heat. Hence they are frequently termed heat rays. 
 It is therefore very necessary to guard the eyes from these rays, and, 
 
254 
 
 MODERN METHODS OF WELDING 
 
 although they are absorbed to a certain extent by ordinary coloured 
 glass, this is not sufficient protection against any intense source. 
 There are, however, several kinds of glass which, although fairly 
 transparent to visible light, are wonderfully efficient in absorbing 
 heat. 
 
 Corning glass G 124 J is one of those which, while it transmits 
 60 to 70 per cent, visible light, cuts off about 90 per cent, of the heat- 
 rays . The colour of this glass 
 is pale green. The author has 
 a pair of goggles fitted with 
 plain lenses of this glass and 
 has found them invaluable 
 when operating on high-tem- 
 perature work. There are, 
 also, gold-fitted glasses which 
 are superlatively efficient 
 in absorbing and reflecting 
 the infra-red heat rays. A 
 sample of the " pfund gold 
 glass " previously referred to 
 was found by careful test to 
 transmit only 0-8 per cent, 
 of the heat rays generated by 
 a 200-watt gas-filled tung- 
 sten lamp operated at normal 
 voltage, the temperature of 
 the tungsten spirals being 
 estimated at 2,400° C. This 
 glass transmits light of a 
 green colour and is much 
 darker than the corning G 
 124 J, so it probably passes 
 not more than 20 per cent, of 
 the visible rays. The novi- 
 weld glasses, especially those of dark tints, are also very efficient 
 shields against the infra-red rays. The effects of even low-power 
 heat rays, when generated in close proximity to the eyes for a 
 considerable time, are often serious, as is evidenced by the fact 
 that glass-blowers who use their unprotected eyes near to hot 
 gas flames of weak luminous intensity, are frequently afflicted with 
 cataract, which might be positively avoided by wearing spectacles 
 made with plain lenses of the G 124 J glass or its equivalent. 
 
 Fig. 128.— Goggles: 47 H, 48 H, 49 H. 
 
EYE -PROTECTION IN IRON WELDING OPERATIONS 255 
 
 Table I. indicates roughly the percentage of heat rays transmitted 
 by various coloured glasses of given thickness. The source of heat 
 used was a 200-watt gas-filled Mazda lamp operating at a temperature 
 of about 2,400° C. Although substantially correct for the samples 
 tested, they would necessarily vary somewhat for other samples 
 of different thickness and degrees of coloration, so that they 
 can be taken only as a general guide for comparative purposes. 
 
 Table I. 
 
 Kind of Glass. 
 
 Thickness in Inches. 
 
 Per Cent. Heat Rays 
 Transmitted. 
 
 Clear white mica 
 
 0-004 
 
 81 
 
 Clear window glass 
 
 
 
 
 0-102 
 
 74 
 
 Flashed ruby 
 
 
 
 
 0-097 
 
 69 
 
 Belgian pot-yellow 
 
 
 
 
 0-126 
 
 50 
 
 Cobalt-blue 
 
 
 
 
 0-093 
 
 43 
 
 Emerald-green . . 
 
 
 
 
 0-1 
 
 36 
 
 Dark mica 
 
 
 
 
 0-007 
 
 15 
 
 Corning G 124 J glass . 
 
 
 
 
 0-095 
 
 10 
 
 Dark noviweld . . 
 
 
 
 
 0-096 
 
 4 
 
 Pfund gold plated 
 
 
 
 
 0-114 
 
 0-8 
 
 We now come to the invisible ultra-violet rays, which are princi- 
 pally to be feared, not only because they are invisible, but because, 
 as previously stated, we have no organ or sense for detecting them, 
 and we can only trace their existence by their effects. In all cases, 
 however, when we are forewarned of their presence they are very 
 easily shielded, for there are only a few substances which are trans- 
 parent both to the visible light and to ultra-violet radiation. Fore- 
 most among these latter substances, because it is most common, is 
 clear natural quartz, or rock crystal, from which the so-called 
 " pebble " spectacle lenses are made. 
 
 Fluorite and selenite are also transparent to ultra-violet rays, 
 but these crystalline minerals are rare and not in common use. 
 However, a moderate thickness of ordinary clear glass, sheets 
 of clear or amber mica, are opaque to these dangerous rays. As a 
 case in point, it is well known that the mercury vapour lamp, when 
 made with a quartz tube, is an exceedingly dangerous light to the 
 eye, being a prolific source of ultra-violet radiation, so that when it is 
 used for illumination it is always carefully enclosed in an outer globe 
 of glass ; when the mercury vapour lamp, however, is made with 
 a clear glass tube, it is a harmless if not very agreeable source of 
 light, because the outer tube of clear glass is opaque to the ultra- 
 
256 MODERN METHODS OF WELDING 
 
 violet rays that are generated abundantly within it by the highly 
 luminous mercury vapour. 
 
 When operating with a source of light which is known to be rich 
 in ultra-violet rays, such as the iron arc in welding operations, it is 
 not sufficient to guard the eyes with ordinary spectacles, because 
 these invisible rays are capable of reflection just the same as visible 
 light, and injury may easily ensue from slanting reflections reaching 
 the eyes behind the spectacle lenses. Goggles that fit closely around 
 the eyes are the only sure protection in such cases. Also, when 
 using a hand shield, such as that shown in Fig. 124, the shield should 
 be held close against the face and not several inches from it. 
 
 It may here be mentioned that the ultra-violet rays, when thej 
 are not masked or overpowered by intense visible light, produce 
 the curious visible effect termed "fluorescence" in many natural 
 and artificial compounds — that is, these rays cause certain com- 
 pounds to shine with various bright characteristic colours, when 
 by visible light alone they may appear pure white, or of some weak 
 neutral tint. Thus, natural willemite, or zinc silicate, from certain 
 localities (which may also be made artificially) shows a bright green 
 colour under the light from a disruptive spark between iron terminals, 
 whereas this compound is white, or nearly so, by visible light. Also, 
 all compounds of salicylic acid, such as the sodium salicylate tablets 
 which may be bought at any drug store, are pure white when seen 
 by visible light, but show a beautiful blue fluorescence under ultra- 
 violet rays. Many other chemical compounds could be mentioned 
 which possess this curious property, but the above substances will 
 suffice to illustrate the effect of fluorescence produced by ultra- 
 violet rays and by which these rays may be detected. It must, 
 however, be noted that these substances will only show their fluores- 
 cent colours very faintly when viewed by the light of low-tension iron 
 arc used in welding, because the intense light of this arc will over- 
 power the weaker effect of the invisible ultra-violet rays. The true 
 beauty of fluorescent colours can only be seen under a high-tension 
 disruptive discharge between iron terminals, the invisible light in 
 this case being weak while the ultra-violet rays are comparatively 
 intense. 
 
 Summarising the effects of means for eye-protection against 
 various harmful radiations, particularly associated with welding 
 operations : 
 
 (1) The intense glare and flickering of the visible rays should be 
 softened and toned down by suitable coloured glasses selected by 
 an expert and having a depth of coloration which shows the clearest 
 
EYE-PROTECTION IN IRON WELDING OPERATIONS 257 
 
 definition combined with sufficient obscuration of glare, which last 
 feature can best be determined by the individual operator. 
 
 (2) When infra-red rays are present to a dangerous degree a 
 tested heat-absorbing or heat-reflecting glass should be employed, 
 either in combination with a suitable dark-coloured glass, when a 
 glaring visible light is present, or by itself in cases where the visible 
 rays are not injuriously intense. 
 
 (3) In guarding the eye from dangerous ultra-violet rays it must 
 be noted carefully that "peb- 
 ble " lenses are made from 
 clear quartz, or natural rock 
 crystal, and this material, be- 
 ing transparent to these rays, 
 offers no protection against 
 their harmful features. On 
 the other hand, ordinary 
 clear glass is a protection 
 against these rays when they 
 are not very intense, but 
 dark amber or dark amber- 
 green glasses are absolutely 
 protective. Glasses showing 
 blue or violet tints should 
 be avoided except in certain 
 combinations wherein they 
 may be used to obscure other 
 colours. 
 
 No. 47 H goggles are light, 
 rust-proof, sanitary, strong, 
 and perfectly fitting. They are fitted with essentialite amber- 
 coloured lenses which afford full protection to the eyes and cover- 
 glasses which protect coloured lenses. 
 
 No. 48 H goggles have the wire shield and other metal parts 
 covered with chamois ; nose-piece is soft leather. 
 
 No. 49 H goggles have aluminium frame, are very light in 
 weight, and are exceedingly popular with welders. 
 
 No. 65 H spectacles have a nickelled steel frame and flexible 
 ear-holds. Fitted with amber lenses, they are very light and com- 
 fortable to wear. 
 
 No. 76 H spectacles are fitted with essentialite amber lenses, 
 have flexible ear-holds and light fibre frame, making them very 
 popular with welders doing light work. 
 
 17 
 
 Fig. 129.— Goggles: 65 H, 76 H. 
 
CHAPTER XXXIX 
 MIRROR WELDING 
 
 Mirror welding is used in some of those rare cases where breaks and 
 fracture occur in inaccessible places, in which ordinary methods 
 could not be adopted. This method is quite a new one, and not 
 known to many, and it is useful in many cases where dismantling 
 would have to take place in ordinary welding. It is used where the 
 space between the article for welding and the obstructing surface 
 is too small. 
 
 The operator is not able to get between the two surfaces, nor 
 could the article be turned. Sometimes two or three mirrors are 
 used together, and set at different angles, so as to facilitate ease in 
 welding, and this adjustment has to be made very accurately so that 
 the operator can do the welding with ease, with the welding line 
 always in view. It is very important that the operator has every- 
 thing in proper order and in the exact position for welding, 'so that 
 during the welding there should be no stopping. 
 
 It may occur in some instances where the welding has to be done 
 in higher places than ordinary ones. Precautions must therefore 
 be taken to secure stability of the structure used. 
 
 In most cases of mirror welding, dissolved acetylene compressed 
 in cylinders (the same as oxygen) is used. Usually these repairs 
 are far from the factory, and in places where an acetylene generator 
 would not be allowed. In case the dissolved acetylene is used, an 
 acetylene regulator would be required for the acetylene cylinder. 
 These acetylene regulators have a left-hand screw at the coupling, 
 and the regulator is painted red, so as to distinguish them from the 
 oxygen regulators (painted black). 
 
 The operator must study carefully the whole job that he has 
 in hand, seeing whether it is necessary to preheat the article and 
 to guard against unequal and invisible stresses and strains. See 
 if the metal is \ inch thick or over ; if so, it must be bevelled : in the 
 instance which we are referring to it would be difficult to bevel. 
 Hence, if a sound weld is to be made the bevelling must be accom- 
 plished. It can be done by the cutting blowpipe, which, if the handle 
 
 258 
 
MIRROR WELDING 
 
 259 
 
 of the cutter is held parallel to the pipe now being welded, and the 
 cutter head pointed at 45 degrees to the point of welding, will 
 bevel one side of the line of welding; the cutter should now be taken 
 to the opposite side and the operation repeated. This should be 
 clear of oxide before starting to weld. 
 
 In welds of this description there must be two operators, one each 
 
 Fig. 
 
 130. — Showing the Principle of Mirror Welding with Specially 
 Arranged Filler Rod. 
 
 side, one using the blowpipe and the other the welding -rod. Special 
 care must be taken by both operators in the finding of the correct 
 point on the line of welding through the mirrors, and must not, under 
 any circumstances, withdraw their visions from the mirrors until 
 the welding line has been completed. 
 
 A smaller blowpipe than usual should be used, as in all vertical 
 and overhead welding the melted metal must not get overheated 
 
260 
 
 MODERN METHODS OF WELDING 
 
 or it will become too fluid, will not adhere, will fall from the weld, 
 and the metal will be burnt and cause a larger space to be filled up, 
 and it would be oxidised, burnt, and cinderised. 
 
 All that is necessary is to heat as small a surface as possible, not 
 more than J inch from the bevel (only heat the outer surface to 
 about 1,000° C.J. Before starting to do any welding it is necessary 
 
 Fig. 131. — Mirror as Applied to the Pipe heretofore Described. 
 
 to see that all the equipment is in perfect order; a lighted torch 
 should be tried to see for certain that it is correct for proceeding. 
 
 The welding should be commenced at \ inch below the break or 
 fracture : this will ensure that the break or fracture will not extend 
 farther. Assuming that one has got all correct, and the trial is 
 satisfactory, welding should now go forward, remembering that one 
 must have a perfect neutral flame, neither oxidising nor carbonising, 
 and the flame must be kept up for certai i while welding. An 
 
MIRROR WELDING 
 
 261 
 
 oxidising name causes adhesion, lack of penetration, oxidation, 
 burnt and cindered weld, and the tests will fail. 
 
 The welding-rod must be absolutely pure, free from phosphorus, 
 sulphur, manganese, and other impurities; the size must be deter- 
 mined by the thickness of the metal to be welded ; several rods should 
 
 Fig. 132.— Internal Welding oe a Boiler, with a Mirror. 
 
 be kept at hand before starting welding. Welding may now be 
 started; the blowpipe must be kept on the particular welding line 
 until such time as the bottom is melted, when the bottom is found 
 at the starting-point; there should be no mistake about the line 
 being continued from the bottom of the weld. 
 
 The welding must be homogeneous, starting at the bottom as 
 
262 MODERN METHODS OF WELDING 
 
 previously stated, and as soon as this is melted add welding-rod, 
 previously heated, in the bevel and move the blowpipe forward 
 with an elliptical sweep, keeping it close in the line of the welding ; 
 do not let the white tip touch the metal, but keep it § inch from it, 
 and go steadily forward, filling up the bevel uniformly with the feed- 
 ing-rod until the end of the weld is reached. There must be no 
 stoppage whatever, while welding the line fractured. If it is 
 done quickly and filled in as the welding proceeds, with no stoppage, 
 the weld will be a success — neat and strong. 
 
 As soon as the welding is completed it is necessary that it be 
 heated to 950° C. and allowed to cool slowly, free from air. 
 
 The mirror welding of a boiler, as shown in Fig. 132, is one 
 that needs every care and consideration; more so than the pipe 
 job previously referred to, because a boiler has to stand very 
 severe tests and strains during its working under steam. Another 
 important point in these boiler cases is expansion and contrac- 
 tion and the avoiding of internal and invisible strains. In the 
 welding of this boiler it must first be made thoroughly clean; all 
 deposited scale that has accumulated must be removed from the 
 fracture and surroundings, and the line of welding must be filed 
 to remove all rust, leaving it bright and smooth. 
 
 Before welding can take place it is necessary to preheat a large 
 area of the boiler internally so that the expansion and contraction 
 may be spread over a larger area than the small confines of the weld. 
 In this case it would be difficult to bevel the edges of the fracture; 
 with the cutting blowpipe, therefore, in place of the bevelling, a size 
 larger blowpipe may be used to penetrate right through the metal. 
 
 Before preheating, it is necessary to have all equipment ready, 
 the mirrors fixed temporarily and marked at the proper angle, and 
 the blowpipe tried in position for welding. This preheating should 
 be carried out by putting a fire inside the boiler until it reaches the 
 temperature of 950° C. When this temperature is reached, immedi- 
 ately put back the mirrors in place to the angle previously marked 
 and commence welding without delay, and see that the temperature 
 does not get below 800° C, or cracks or fractures will take place. 
 
 The welding should be started at | inch beyond the line of weld- 
 ing so as to prevent the crack or fracture extending farther than the 
 present line of welding. It is not necessary to have two operators 
 on a job like this. The blowpipe is to be one size larger than is 
 usual for the thickness of metal being welded, and the pressure of 
 oxygen to be slightly less than that stated on the blowpipe. First, 
 the surroundings of the crack or fracture should be heated to about 
 
MIRROR WELDING 263 
 
 1,000° C.j a little above the preheating temperature, and then start 
 welding at the point previously referred to ; this point takes more 
 heating than the other part of the welding line. 
 
 There are difficulties of lack of penetration, bad joining, blow- 
 holes, and interposition of oxide. Lack of penetration is a frequent 
 occurrence. There is, however, no justification for this, if operators 
 will only go to the bottom of the weld in all cases. Interposition 
 of oxide is a common occurrence, and all operators should study 
 this and make test pieces until they are satisfied that there is no 
 interposition. It occurs, chiefly, with excess of oxygen and using 
 too large a blowpipe. The oxide formed through these errors is im- 
 prisoned in the metal. It is very important to see that during the 
 welding no adhesion takes place and full penetration has been 
 observed, that there is no oxidation nor blowholes, and that 
 no part of the weld has been gone over twice without adding 
 the welding-rod. When the welding is completed, the temperature 
 should be immediately taken, and whatever it is it must be raised 
 to 950° C. by putting a fire in the boiler, then allowed to cool slowly, 
 free from air. When cold, test by hydraulic pressure to double 
 the working pressure. 
 
 Assuming that welding is now starting, the operator must start 
 just below the fracture line, get well hot over an area of 3 x 3 inches 
 before melting at the starting-point — this will increase the speed of 
 welding — penetrate fully, keeping the tip of the blowpipe vertical 
 and at an angle of 40 degrees, which will just suit the fracture, go 
 along slowly and regularly with a gyratory movement, adding the 
 necessary welding-rod, filling the fracture to its greatest extent with 
 a little extra coating to give more strength ; be sure that there is no 
 stoppage ; the molten metal must be semiplastic and not cinderised 
 in any way, or too fluid — just at a temperature that will scarcely 
 run; a good weld will then be obtained. 
 
 
 PRINTED IN GREAT BRITAIN BY 
 BILLING AND SONS, LTD., GUILDFORD AND ESHER 
 
are prepared to supply, either from 
 
 their complete stock or at 
 
 short notice, 
 
 Any Technical or 
 
 Scientific Book 
 
 In addition to publishing a very large 
 and varied number of Scientific and 
 Engineering Books, D. Van Nostrand 
 Company have on hand the largest 
 assortment in the United States of such 
 books issued by American and foreign 
 . publishers. 
 
 All inquiries are cheerfully and care- 
 fully answered and complete catalogs 
 sent free on request. 
 
 8 Warren Street - - New York 
 
306 90 
 

 & ^ o 
 
 vr«v 
 
 
 
 
 
 
 „•/ V'^S-v* "V^^V* ^v^i^v* 
 
 °*'*'~»' A ' V 
 
 
 4> 
 
 *° 1 
 
 
 
 iP**. 
 
 «?-«» 
 
 r^. 
 
O^. 'o , » 
 
 
 
 t ' c 
 
 
 *6* 
 
 ° J*